fM^i .v.^f&v^miiff.i Bibliotheque botanique EMILE BURNAT 4'a(uliM|iie \" Provicnl Jc LiM-cs proNfiuinl iU' hi l»ilili()lli<''<|u«' li<>l;iiii(|iic (rEmile Burnatt l8-28-l*)-20:), iiisrivs en oclohiv l'.)-2U (in llii yt ^^^:'k. K r: THE JOURNAL OF BOTANY BRITISH AND FOREIGN. KDITKU BY JAMES BRITTEN, K.S.G., F.L.S L\TR Senior Assistant, Department of Botany, British Museum. tWRART pew YORK UOTANiCAL UAKOBM VOL. LII WITH PLATES AND TEXT ILLUSTRATIONS LONDON WEST, NEWMAN .fe CO., 54, HATTON GARDEN. 19 14. LONUON : PKINTED BY WEST, NEWMAN AND CO. HATTON GARDEN, E.C. CONTEIBUTORS TO THE PRESENT VOLUME. E. G. Baker, F.L.S. J. G. Baker, F.L.S. W. Barclay. E. J. Bedford. Arthur Bennett, A.L.S. Spencer H. Bickham, F.L.S. V. H. Blackman, F.R.S. Sidney F. Blake, A.M. G. S. BOULGER, F.L.S. James Britten, F.L.S. W. H. BURRELL, F.L.S. F. Cavers, D.Sc, F.L.S. F. J. Chittenden, F.L.S. Miller Christy, F.L.S. J. E. Cooper. A. D. Cotton, F.L.S. H. N. Dixon, M.A., F.L.S. R. A. DUMMER. J. B. Farmer, F.R.S. W. Fawcett, B.Sc, F.L.S. Antony Gepp, M.A., F.L.S. Col. M. J. GODFBRY. G. GooDE, M.A. F. J. Hanbury, F.L.S. J. W. Hartley. W. P. HiERN, M.A., F.R.S. E. M. Holmes, F.L.S. A. Bruce Jackson. B. Daydon Jackson, Ph.D., Sec. L.S. C. E. Larter, F.L.S. G. Lister, F.L.S. A. S. Marsh, B.A. B. S. Marshall, M.A., F.L.S. Arthur Mayfield. J. Cosmo Melvill, D.Sc, F.L.S. A. H. Moore. Spencer le M. Moore, F.L.S. C. E. Moss, D.Sc, F.L.S. Seiitchi Narita. W. E. Nicholson. R. Paulson, F.L.S. S. R. Price. H. W. PUGSLEY, B.A. J. Ramsbottom, B.A., F.L.S. J. K. Ramsbottom. Clement Reid, F.R.S. E. M. Reid, D.Sc, F.L.S. A. B. Rendle, D.Sc, F.R.S. H. J. Riddelsdell, M.A. H. N. Ridley, C.M.G., F.R.S. W. D. Roebuck, F.L.S. W. Moyle Rogers, F.L.S. C. E. Salmon, F.L.S. A. LoRRAiN Smith, F.L.S. F. Stratton, F.L.S. E. N. Thomas, D.Sc. H. H. Thomas. H. S. Thompson, F.L.S. W. G. Travis. T. Wainwright. H. F. Wernham, D.Sc. F.L.S. G. S. West, D.Sc, F.L.S. J. A. Wheldon, F.L.S. J. W. White, F.L.S. F. N. Williams, F.L.S. W. Wilson. F. G. Wiltshear. E. A.Woodruffe-Peacock,F.L.S. Major A. H. Wolley-Dod. Directions to Binder. Tab. 529 to face page 1 „ 530 89 „ 531 113 „ 532 193 „ 533 225 „ 534 257 Portrait OF William West. 161 The Supplement (' A Flora of Gibraltar and the Neighbourhood should be placed separately at the end of the volume. Journ.Bob. Tab. 529. P. Higl-ley del.et Jith . West.Newman imp . Tal'botiella eketensis Bak.fil. I IN. C3 LIKRAKY PEW YORK. BOTANICAL THE QAltOBN JOURNAL OF BOTANY BRITISH AND FOREIGN. PLANTS FROM THE EKET DISTRICT, S. NIGERIA COLLECTED BY Mr. AND MrS. P. AmAURY TaLBOT. (Plate 529.) Mr. and Mrs. P. Amaury Talbot have been continuing tlieir work of botanical exploration in Southern Nigeria. For the last twelve months Mr. Talbot has been stationed in the Eket District, a broad strip of land bordering the Gulf of Guinea, stretching westward from Calabar and the Cross River. Mr. Talbot de- scribes the land bordering the shore of the Gulf as one vast littoral, crossed and recrossed by a network of waterways, so that it is possible to pass by canoe from French Dahomey on the one side to the German Cameroons on the other without once sighting the sea. The country is drained by the inner stretches of the Cross River and the Kwa Ibo. It consists of mangrove marsh lining the banks of creek and stream, and of fertile palm swamp with coco-nut, piassava, and oil-palms, while toward the sea- shore dwarf dates fringe the low-lying lands. Unfortunately, owing to the unsettled nature of the country, botanical exploration has been carried on with some difficulty, but notwithstanding this, a large collection has been sent to the National Herbarium, where it is in course of determination. Many cvj specimens are identical with those previously sent by the Talbots from the high-lying Oban District, but the collection contains a good proportion of West African species not yet recorded from Nigeria, though generally previously known from the Cameroons. An interesting feature is the presence of species hitherto known only from material collected by Gustav Mann at Calabar. There are also a good number of novelties, and it is proposed to publish descriptions of these, and notes on other species of interest, in the pages of this Journal. The present instalment includes a new genus of Leguminosae and a number of new species, especially of the families Rubiaceae and Apocynaceae. Mrs. Talbot has paid special attention to the Napoleonas, and the collection will supply a substantial addition to the number of species hitherto known from West Tropical Africa. The notes as to habit, &g., have been kindly supplied by Mrs. Talbot. A TJ "R Journal of Botany. — Vol. 52. [January, 1914.] b 2 THE JOURNAL OF BOTANY POLYPETALiE. By E. G. Baker, F.L.S. Anonace^. Isolona campanulata Engler & Diels, No. 3261. — An interesting Cameroons species, not previously collected in South Nigeria. Sepals bright green petals, centre blackish purple. Eket-Ibeno Eoacl. Bixine^. Oncoba Mannii Oliver. — This seems to be rather common in Southern Nigeria and also at Bipinde in the Cameroons. Zenker, Nos. 1637-2860-2333 distributed as 0. aristata Oliver, belong here. Legubiinos^. Talbotiella, Baker fil., nov. gen. Beceptaculum breviter infundibuliforme. Sepala 4, suborbicu- laria, aestivatione late imbricata, inter se fere aequalia. Petala 0. Stamina 8-10 inter se parum inaequalia, filamentis filiformibus inter se liberis, antheris versatilibus, longitudinaliter dehiscenti- bus. Ovarium stipitatum, villosum, 2-ovulatum, stipite recepta- culo adnata, stylo terminali elongato, sursum glabro, stigmate parvo, capitellato. Legumen ignotum. Frutex foliis paripinnatis foliolis multis oppositis. Bacemi axillares, laxe pluri vel multiflori. Bracteolcs 2, submembranaceae, lineari-oblongge, baud involucrem formantes, apicem versus pedi- cellorum positas. Squam(B ad basin pedunculorum scariosae, brunnescentes, imbricatse, cymbiformes, dorso nitidae. T. eketensis, sp. nov. (PI. 529). — Bamuli lignosi cortice nigrescente obtecti ; stiimlis angustis ; foliis multifoliolatis, /oZioZw oppositis 12-16 jugis oblongis sessilibus, obliquis (margine postico basi in triangulum acutum brevissimum protracto) apice rotun- datis, superne glabris, subtus breviter pubescentibus, costa sub- centrali, rhachide tenui superne canaliculata ; florihus in racemes pluri vel multiflores et laxos dispositis, pcdicellis gracilibus pubescentibus, bracteoUs lineari-oblongis, sepalis ad basin fissis membranaceis concavis : receptaculo extus pilis sparsis obtecto ; fetalis nullis ; staminibtis 8-10 filamentis filiformibus calyce longioribus ; ovario stipitato villoso, stylo tenui, stigmate parvo ; fructu ignoto. Ibeno, at estuary of Kwa Ibo Eiver, No. 3188. Bush 1-6 metres. Leaves 5-8 cm. long, leaflets 7-13 mm. long, 3-4-5 mm. broad. Bracteoles + 8 mm. long. Calyx 4-5 mm. long. Keceptacle + 2 mm. long. Ovary 2-5 mm. long. This genus belongs to the tribe Cynometreae of the Cffisalpineae, and is allied to Cynometra, especially to such species as C. Hankei Harms. The genus Cynometra, as defined in Bentham & Hooker's Gen. Plant., consists of trees or shrubs, with paripinnate leaves in one or few pairs, five petals, and an arcuate-ovoid or subreniform, rarely straight, thick, turgid, rarely subcompressed pod. It there- PLANTS FROM THE EKET DISTRICT, S. NIGERIA 6 fore seems undesirable to place in Cynometra plants possessing the structure of Talhotiella. T. eketensis is very closely allied to the plant described by Dr. Harms (in Engl. Jahrb. xxvi. 267) as CryptosepaUtm ? Staudtii. Unfortunately good flowers were not present when his species was described, and the number of the stamens is doubtfully considered to be 2-3. Talhotiella is quite different in structure from Crypto- sepaluvi ; in this latter genus the calyx, as the name implies, is minute, there is one posterior petal, three stamens, and the ovary is 2-4-ovulate. Mrs. Talbot's notes are as follows : — Bushes growing in pro- fusion in rather loose drift sand. Highest growth 4-6 metres. Dwarfy shrubs flower at 1 metre. New leaves very pale green shaded mauvy pink, older leaves dark glossy green. Flowers all white ; stamens and anthers brightest orange ; petiole white at top, palest green below. Bracts bright pink. EOSACE^. Strepho7iema Mannii Hook, fil. Agrees with Mann, No. 2293 in Hb. Kew, from Old Calabar and Gaboon rivers. The position of the genus Strephonema is a matter of some uncertainty. Bentham & Hooker in their Genera Plantarum place it in Lythracese, from which Koehne •= excludes it in his Mono- graph of the order. Baillon has suggested that it is an abnormal Eosaceous plant. The ovary is partly adherent to the tube of the calyx, and there are two pendulous collateral ovules. Mr. Spencer Moore and I, who have both dissected flowers independently, came to the same conclusion, and incline to agree with Baillon's suggestion, and consider that the structure indicates considerable affinity with such genera as Pygeum in the Pruneae tribe of Eosacege. Ehizophorbje. Cassipourea eketensis Baker fil., sp. nov. Bamuli graciles teretes glabri ; foliis inter minores generis oppositis ovatis basi cuneatis breviter petiolatis subcoriaceis glabris apice acuminatis apice ipso obtusis nervis lateralibus utrinque 5-7 ante marginem arcuatis inter sese conjungentibus costa subtus prominente ; stipulis caducis ; florihus axillaribus glomeratis ; pedicellis calyce brevioribus ; calyce campanulato usque ad medium 5-lobato lobis triangularibus acutis erectis valvatis ; petalis unguiculatis apice fimbriato-laceratis ; staminibus calyce duplo longioribus antheris subglobosis; s^?//o tereti stigmate dilatato; ovario triloculari ovulis in loculis 2 axi collateraliter affixis. Sine numero. Leaves ovate, acuminate, the actual apex being obtuse, 5'6cm. long, 16-27 mm. broad, petiole 1-2 mm. long. Calyx + 5 mm. long. Petals fimbriate-lacerate, + 9 mm. long. Stamens nearly twice as long as the calyx. Disc cupuliform. Ovary trilocular, hairy. * See also Sitzungsbericht Bot. Verein. Prov. Brand, xxii. 66. B 2 4 THE JOURNAL OF BOTANY Allied to Weihea Afzelii Oliver, but this plant has smaller leaves and more numerous flowers. No. 3234 is closely allied but the leaves are longer. Passiflore^. Soyauxia Talbotii Baker fil., sp. nov. Frutex 5-6 metr. (Mrs. Talbot). Bamuli sursum parce strigoso-pilosi deorsum fere glabri ; stipulis mox deciduis ; foliis subcoriaceis petiolo brevi instructis oblongis basi acutis apice acuminatis nervis lateralibus circ. 12-13 patentibus inter sese per venas tenues laxe reticulatas conjunctis; inflorescentiis axillaribv;s laxiuscule racemosis rhachide fulvo-tomentosa ; florih^is breviter pedicellatis ; bractcis parvis ; calycis segmentis ovatis extus ferrugineo-tomentosis ; ijetalis ob- ovatis obtusis sepala sequantibus ; filamentis filiformibus petalis longioribus ; disco tubiformi brevi ; ovario piloso ; stylis filiformi- bus subulatis sepala superantibus. Ikotobo Eoad. No. 3254. Leaves 14-16"5 cm. long, 45-48 mm. broad, lateral nerves 10-20 mm. apart. Petiole + 3 mm. long. Eacemes 6-0-10'Ocm. long. Pedicels + 1 mm. long. Sepals 3-5-4-0 mm. long. Styles + 7 mm. long. Very closely allied to S. hipindensis Gilg., but the flowers are slightly larger and the lateral nerves of the leaves are more distant. Flower white, with dark very small anthers. GAMOPETALtE. By Spencer Moore & H. F. Wernham. KUBIACE^. Urophyllmn eketense Wernham, sp. nov. Frutex erectus, ramulis novellis valde compressis viridibus, tardius glabrescentibus subteretibus cortice laivi brunneo ; foliis magis pergamaceis ob- ovato-oblongis vix acuminatis, utrinque nisi subtus in venis obscure puberulis glabris supra nitentibus, venis conspicuis supra impressis subtus prominentibus, basin versus in petiolum minute puberulum breviusculum leniter angustatis, stipiilis ovatis majusculis viridis- simis appresse puberulis ; florihus in cymis axillaribus nee quasi- terminalibus paniculatis minute puberulis multifloris laxiusculis pedunculatis, bracteis linearibus subfiliformibus; calyce subintegro; bacca pisiformi glaberrimo, 4-loculari. Oron-Eket main road. No. 3327. The leaves average from 18 cm. x 6 cm. to over 21 cm. x 8 cm. ; with stalk from 2-3 cm. long ; lateral nerves, about 20 on each side of the midrib. Stijndes, 1'5 cm. x 8 mm. Fruiting p)eduncles, 2-2-5 cm. long. This species has evident affinities with the Central African U. viridiflorum Schweinf., but it is distinct in the shape of its leaves and their longer stalks, and in the wholly axillary inflorescence. Tarenna eketensis Wernham, sp. nov. Arbor ca. 30-pedalis, ramulis divaricatis, novellis strigillosis ; foliis chartaceis ellipticis . PLANTS FROM THE EKET DISTRfCT, S. NIGERIA 5 utrinque angustatis breviter acuminatis obtusis, venis secundariis subtus prominentibus appresse sericeo-pubescentibus, reticulo tertiario conspicuo plus minus impresso, petiolo brevi tamen manifesto, stipulis ovato-triangularibus glaberrimis apice rotun- datis ; infloresccntia laxiuscula, raraulis qua pedicelli et ovaria densiuscule appresse pubescentibus demum glabrescentibus, bracteis foliaribus vel parvis ; calycis dentibus brevibus acutis ; coroUcB tubo extus strigilloso, lobis glabris oblongis obtusis tubi dimidium excedentibus. Main road from Oron to Eket, mostly in farm-clearings. No. 3024. Near T. nitidula Hiern, differing especially in the thicker leaves with their distinctive venation, the obtuse stipules, the pubescent corolla, &c. The leaves attain about 12 cm. x 5 cm., with stalk barely 8 mm. long ; there are 6-7 secondary nerves on either side of the midrib. Stipules, about 7 mm. x 4 mm. at most. Primary lateral peduncles, 3-4-5 cm. long ; thyrsus about 4 cm. long and 6 cm. across. Pedicels, 4-5 mm. long. Corolla- tube 4-5 mm., lobes 3 mm. Anthers 4 mm. long ; style exserted over 5 mm. The ultimate inflorescence-branches, pedicels, ovaries, and calyces are densely clothed with a minute but conspicuous golden- brown silky pubescence, and the youngest branchlets are more or less sparsely strigose, as well as the leaf- veins on the under sur- face ; otherwise the plant is glabrous, with shining leaves. " Calyx-lobes pinkish ; corolla-lobes greenish-white ; stamens cream ; pistil white." Gardenia Cunliffeae Wernham, sp. nov. Frutex scan dens glaber, in siccitate omnino nigricans, ramulis novellis brevissimis, tardius cortice brunneo rugosulo indutis ; foUis glabris obovato- oblongis apice rotundato, petiolatis, stipulis parvis ovato-lanceo- latis acutis caducis ; florum fragrantissimorum cymis sessilibus saepius 3-floris glaberrimis, pedicellis breviusculis, in ovarium et calycem limbo dentibus brevibus 5 acutissimis deciduis insensim dilatatis ; corolla tubi parte brevi inferiore anguste tubulari dilute viride insuper subito late infundibulariter dilatato, lobis candidis purpureo-maculatis late oblongis apice rotundato. Ubium Eiver. No. 3149. A plant preserved in the Kew Herbarium, collected in Lagos by Millen (No. 144), is referable to the same species. The leaves attain a size of over 13 cm. x 7 cm., with petiole over 2 cm. long ; there are 4-5 pairs of secondary veins ; stipules 8 mm. X 5 mm. Pedicels 5 mm. or longer. Calyx and ovary 3 mm. Lower, tubular part of corolla projects about 5 mm. beyond the calyx ; the upper, funnel-shaped part of corolla-tube measures 2-5 cm. in length, and 3-5 cm. wide at the mouth ; co?'o//a-lobes 1-7 cm. x 1'4 cm. The nearest affinity seems to be G. Annce E. P. Wright, but this is readily distinguished by the shape of the leaves alone. " Corolla-tube very pale-green externally, dark green inside toward the base ; lobes white above, with vivid purple splashes. O THE JOURNAL OF BOTANY Stamens cream. Stigma yellow, with cream tip. Fruit like a small plum, vivid orange" (Mrs. Talbot). Named after Mrs. Cunliffe in recognition of her keen and practical interest in the Nigerian flora. Randia Galtonii Wernham, sp. nov. Frutex ramulis tereti- bus viridibus molliter et densiuscule pubescentibus ; foliis mem- branaceis 3-natis oblanceolatis basi cuneatis longiuscule saepius caudato-acuminatis acutissimis, petiolo brevissimo, utrinque proe- cipue subtus et in venis inconspicuis hispidulo-pubescentibus, stijJuUs lato a basi in setis 2 productis ; floribns in axillis solitariis 7-8-meris sessilibus magnis, calycis omnino hispiduli lobis setaceo- subulatis elongatis, corollce extus pubescentis tubo gracili desuper glabrescente, insuper sub lobos oblongos caudato-acuminatos infundibuliformiter ampliato. Oron-Eket main road ; fl. Feb. No. 3219. A striking species, allied to B. octomera Bth. & Hk. f., from which it differs in the more lengthily acuminate and more hairy leaves, and the much larger and differently-shaped corolla-lobes. The leaves are in whorls of 3 ; those at the ends of the shoots are small, and very unequal, two being subequal 9-10 cm. x 2*5-3 cm., and the third about 5 cm. x 1'5 cm. The adult leaves are sub- equal, 10-15 cm. long (exclusive of the almost setaceous acumen often over 2 cm. long) x 4'5 cm. wide above the middle ; secondary veins 5-8 pairs ; the petiole is at most barely 5 mm. ; the low^er, entire, part of the sti]_mles is 8 mm. wide and 3-4 mm. deep, and the two setge above reach 1 cm. in length. Calyx-ixxhe and ovary together 2 cm. long, calyx-lobes 3-4-5 cm. Coro?/a-tube— the slender tubular part — 14 cm. long, the funnel-shaped upper part about 3 cm. deep, and 4-4*5 cm. wide at the mouth ; corolla- lobes 7-8 cm. x 1-1*5 cm. Anthers 2 cm. long. Stigma exserted 2*5 cm. The species is named after Major Galton, of Hadzor, Droitwich, whose interest in the Nigerian flora has led him to attempt the introduction of Nigerian trees and shrubs on his English estate. " Calyx very dark green ; corolla-tube bright green, lobes white within and divided on the back longitudinally into a bright green half and a creamish-white half ; stamens light drab ; style and stigma cream " (Mrs. Talbot). Randia Cunliffeae Wernham, sp. nov. Frutex humilis ample patulus, ramulis hispidulo-pubescentibus ; foliis glauco-griseis (Mrs. Talbot) tenuiter chartaceis 3-natis ellipticis v. oblanceolatis breviter nee caudate acuminatis acutis, brevissime petiolatis, utrin- que hispidulis, stipiilis a basi latiusculo triangulari 2-3-setaceo- acuminatis ; florihus in axillis solitariis 8- vel pleio-meris sessilibus magnis ; calycis hispiduli lobis plano-subulatis rigidiusculis vix setaceis ; corolla extus sparsiuscule asperulo-pubescentis tubo validiusculo, insuper parum ampliato, lobis oblanceolatis breviter cuspidato-acurainatis. Oron-Eket main road. No. 3385. Allied to the preceding species, but readily distinguished by the habit, the shape and colouring of the leaves, flat calyx-lobes, PLANTS FROM THE EKET DISTRICT, S. NIGERIA ' and by the shape of the corolla, with its rather broad tube, but little widened above. The leaves, including the ultimate ones, are all subequal (cf. B. Galtonii), about 12 cm. x 4 cm., petiole not more than 7-8 mm. Calyx-lohe^ 3-4 cm. long, and as much as 2-3 mm. broad. Co?*o//a-tube over 14 cm. long, 6-7 mm. wide at the middle when dry, and less than 2 cm. wide at the mouth ; lobes 3"5 cm. X 9 mm. Canthium viridissimum Wernham, sp. nov. Kamis sub- teretibus senioribus cortice ssepius fusco-brunneo conspicue lenti- cellato indutis, novellis brevibus glabris flavis complanatis ; foliis membranaceis glaberrimis in siccitate viridissimis ellipticis utrin- que breviter acurainatis, petiolo brevi, stipulis acutis minutiusculis; inflorescentia sessili, in juventute compacta tandem latiuscula foliis tamen multo breviore, floribus glabris pro genere medio- cribus, alabastro obtuso pedicellis viridis gracilibus fiores multo excedentibus ; calyce infundibulari subintegro ; corolla tubo sub- globose inflato. Oron-Eket road, near Ikotobo. No. 3121, and other un- numbered specimens. Eeadily distinguished by the very thin leaves, 8-9 cm. x 8-5-4 cm., and the short, light yellow young shoots. The straight flowering-branches bear flower-clusters not exceeding 3-4 cm. in diameter at each node. Floral pedicels 7 mm. long. The calyx is 2 mm. deep, and the corolla-tube is exserted 3 mm. beyond it, being 4 mm. wide , corolla-lobes 3 mm. long. " Calyx-lobes bright green ; corolla-lobes bright green, white toward the centre of the flower. Anthers orange-brown. Stigma bright orange with dark green centre-spot" (Mrs. Talbot). Cuviera calycosa Wernham, sp. nov. Arbor 90-pedalis, glabra, in siccitate nigricans, ramulis tenuibus mox cortice cinereo indutis ; foliis pergamaceis ellipticis v. oblongis pro genere parvulis, breviter et anguste acuminatis, obtusis, basi acutis, glabris, petiolo breviusculo, stijndis parvis lanceolatis acuminatis nisi basi lato caducis ; cymis paucifloris dichotomis laxiusculis, bracteis oblongo- lanceolatis obtusis ; calyce magno lobis inaequalibus ovato- lanceolatis acuminatis acutissimis, corollam multo excedente ; corollce tubo late infundibulari-cylindraceo breviusculo, lobis 5 longe setaceo-acuminatis, pilis bine inde paucis longiusculis conspersis ; bacca glaberrima calycis a limbo persistente coronata. Near Esuk Ekkpo Abassi. Fl. & Fr. May. No. 3300. A remarkable species, the nearest afdnity being clearly C. nigrescens Wernham ; the present species is distinct, especially in the very large calyx and small corolla. The leaves measure 10-11 cm, X 4-4-5 cm., with petiole about 1 cm. long ; secondary nerves 5-6 pairs ; stijjules 6-8 cm. long. Peduncle 3 cm. ; cyme 11-12 cm. wide, 5-6 cm. long. Pedicel 5 mm. ; calyx-tuhe minute, lobes 3-3-5 cm. x 4-7 mm. Corolla-tuhe barely 5 mm. long, and nearly as much in average breadth ; lobes, flat part 4-5 mm., setae over twice that length. Berry 1-4 cm. x 1-1 cm. " Youngest flowers white, older ones cream, oldest thin orange. 8 THE JOURlNAL OF BOTANY Centre of flower greenish. Calyx-lobes bright green, with margin and sette white. Seta) of corolla-lobes white ; anthers dark-purplish brown ; style white, stigma pale green " (Mrs. Talbot). Coffea eketensis Wernham, sp. nov. Frutex ramosissima, ramis divaricatis decussatis novellis minute puberulis ; foliis chartaceis ellipticis, utrinque angustatis, brevissime petiolatis, apice subacutis, utrinque ipsis in venis glaberrimis, venis secun- dariis paucis distantibus, petiolo puberulo, siipiUis a basi lato brevissimo setaceo-apiculatis ; florihus in axillis soHtariis prse- cocibus, bracteis exterioribus epicalycem tubularem dentibus lineari-lanceolatis acutissimis acurainatis formantibus, hractcolis lanceolatis acuminatis acutissimis valde concaveis ; calycis minutiusculi lobis late rotundatis 8 ; corollce tubo gracillimo lobis oblanceolatis apice rotundatis. Along the rivers, and 2-3 miles from Oron on the Eket road. No. 3064. Leaves about 5 cm. x 2-5 cm., secondary veins 3-4 pairs ; petiole 2 mm. ; stipules, including the setaceous apiculus, about 4 mm. long. Bracteoles 1 cm. long. Calyx barely 1 nrm. deep. Corolla-tuhe 2-2-3 cm. long, lobes 7 mm. x 2 mm. Alhed to C. jasviinoides Welw., differing principally in the thicker leaves with glabrous and fewer veins, and the less precocious flowers with 5-lobed corolla. " Flowers white, bracts bright-brown, calyx-lobes bright pale- green " (Mrs. Talbot). Cephaelis Talbotii Wernham, sp. nov. Frutex giabrescens, 10-15-pedalis, ramulis complanatis bifariatim ferrugineo-pubes- centibus ; /o/ze's magnis ovalibus v. ellipticis basi acutis, breviuscule acuminatis apice subacutis, utrinque nisi subtus in venis sparse puberulis glabris, venis supra impressis subtus conspicue promin- entibus, petiolatis, stipuUs bipartitis ovatis acuminatis basi ferrugineo-pubescentibus membranaceis ; capitulorum multorum inflorescentia in axillis longepedunculata trichotoma, bracteis pri- mariis paucis lanceolatis foliosis, capitulis parvis, involucri bracteis ovato-lanceolatis interioribus dentatis v. 2-3-fidis; calycis hmbo infundibulari brevissime et insequaliter dentate, dentibus apice rubescentibus ; corolla nisi in ore barbate glabrae tubo subglobose ampliato sub lobos oblongos patentes apice incurvato coriaceo. Oron-Eket main road. No. 3386. The leaves measure about 15-25 cm. x 8-11 cm., with stalk 1-5 cm. long ; secondary nerves about 18 pairs, each half of the stipule 2 cm. x 6 mm. Peduncle 8-9 cm. long; bracts 1-3 cm. X 2-5 mm. Inflorescence — a rather lax cyme of small capitula — about 4-5 cm. in diameter, and 3-5 cm. in length, each head being 8 mm. in diameter. Ovary and calyx-\\mh together barely 1-5 mm. Corolla-iwhe 2 mm., lobes about the same length. The nearest allied species is C. cornuta Hiern, which differs, however, chiefly in its pubescent corolla with differently shaped lobes. The collection includes a Rul)iaceous plant in fruit, which appears to be a Cephaelis. The inflorescence is a trichotomous PLANTS PROM THE EKET DISTRICT, S. NIGERIA 9 umbellate cyme of small heads, each of about 1'8 cm. diameter, and comprising as many as a dozen small, ellipsoidal, longi- tudinally furrowed glabrous berries 5 mm. long and 3'5 mm. broad. Each berry is crowned by the persistent, membranous, reddish calyx-limb 1-5 mm. high, with short, narrow, sub-setaceous teeth. The branches of the inflorescence, like the peduncle (7-8 cm.), are densely rufo-pubescent ; the whole inflorescence measures over 7 cm. across. The reddish, papery bracts are from 6-10 mm. long. The single leaf which the specimen bears is glabrous (except for the puberulous midrib below), obovate narrowed toward the base into a petiole about 1 cm. long, and shortly acuminate with subacute apex ; the blade measures 22 cm. X 10 cm. ; secondary nerves about 20 pairs. I have little doubt that this plaut is a fruiting specimen of C. cornuia Hiern. The latter species (Fl. Trop. Afr. iii. 224) is based on a specimen gathered in Old Calabar by Dr. Robb, and preserved in the National Herbarium. This consists of about ten loose inflorescences in the flowering stage, with a couple of more or less immature leaves. The structure and indumentum of the inflorescence, the bracts, and the calyx, are all identical with those in Talbot's plant, and the leaves of Robb's specimen may be reasonably conceived as representing an earlier stage of the leaf in the plant before us. (To be concluded.) DISTEIBUTION OF UTRICULARIA IN BEITAIN. By Arthur Bennett. In this account I have added the records for the Watsonian counties which have come to my knowledge since the publication of the Supplement to Top. Bot. ed. 2 (Journ. Bot. 1905, Supplement). U. VULGARIS L. 60. Lane. W. Flora. 104. Ebudes N. J. Bot. 1910, 225. 109. Caithness. Miss Lillie sp. U. MINOR L. 37. Worcestershire. Flora 280. 76. Renfrew. Ann. Scot. N. Hist. 1891, 106 89. Perth E. Sturrock in Perth Herb. (var. jjlaty- loba Meister teste Gliick). U. MAJOR Schmidel [neglecta Lehm.). 44. Carmarthen. Hamer sp. 46. Cardigan. Salter, 1906. 59. Lancaster S. Edinb. herb. ! 63. YorkS.W. Cardiff herb. ! 69. Westmorland. Martin- dale ! 79. Selkirk. Marshall sp. 89. Perth E. Sturrock sp. 92. Aberdeen S. Trail sp. 96. Easterness. Ann. Scot. N. Hist. 1911, 171. 102. Ebudes S. McNeill sp. 109. Caithness. Lillie sp. 110. Hebrides. Shoolbred sp. U. INTERMEDIA Hayne. 11. Hants S. Borrer in herb. Edinb. ! 89. Perth E. Four localities, teste Dr. Gliick, 10 THE JOURNAL OF BOTANY 90. Forfar. "Eescobie." Dr. Gliick in litt. 91. Kincardine, 1848. Edinb. herb. ! as minor. Mr. Druce gives 96, 100, 105, in Ex. Club Eeport for 1910, 514. The following counties are re- ported for U. intermedia, but need verification : — 25. Suffolk E. Winch add. 59. Lancaster S. Winch add. 67. Northumberland. Thom- son, herb. Watson. 70. Cumberland. Winch Con- trib. 81. Berwick. Border Flora. 85. Fife. Ann. Scot. N. Hist. 1901, 103. 92. Aberdeen S. N. Flora. 93. Aberdeen N. Trail. 95. Elgin. Druce. 107. Suth. E. Graham Excur. Somerset ? Devon ? U. OCHROLEUCA Hartm. 9. Dorset. Linton sp. 11. Hants S. Mennell herb.! 62. York N. E. Martindale. 69. Westmorland. Fox sp. 72. Dumfries. Corrie sp. 73. Kirkcudbright. Coles sp. 74. Wigton. McAndrew sp. 87. Perth W. Perth herb. ! 88. Perth M. Ewing sp. (3200 ft.). 89. Perth E. Druce. 90. Forfar. Edinb. herb. ! 91. Kincardine. Edinb. herb. ! 96. Easterness. Dixon sp. 97. Westerness. Macvicar sp. 98. Argyll. Marshall sp. 99. Dumbarton. Watt sp. 101. Cantire. Ewing sp. 102. Ebudes S. Somerville sp. 103. Ebudes M. Macvicar sp. 104. Ebudes N. Ewing sp. 105. Eoss W. Salmon herb. ! 106. Eoss E. Mennell sp. 108. Suth. W. Miller sp. 110. Hebrides. Shoolbred sp. 112. Shetland. Beeby sp. U. Bremii Heer. At present it is best not to report any county. That of N. Lancashire is an error, reported by me in Journ. Bot. 1912, 316. Dr. Gliick pronounces the specimens only "minor." It differs from our Surrey minor considerably, the flower being double the size. With the exception of the two southern counties, the distri- bution of U. ochroleuca, as at present known, is decidedly northern in Britain. GIBEALTAE PLANTS. By Major A. H. Wolley-Dod. The following notes and descriptions relate to plants which it seems desirable to treat more fully than would be convenient in my Flora of Gibraltar now being issued as a Supplement to this Journal. I am indebted to Mr. N. E. Brown and Mr. Turrill respectively for the descriptions of Euphorbia gihraltarica and ByncJiospora alba var. pauciseta, and to Dr. Stapf for assistance in that of Alropis iberica. Delphinium peregrinum Linn. Sp. PI. ed. 2, p. 740. Linnaeus described the inner petals of this species as subrotund, and the GIBRALTAR PLANTS 11 specimen in his herbarium has them subcordate at the base of the limb. Most authors, however, have regarded them as eUiptical and more or less narrowed below, and from this, together with the synonymy cited, much confusion has arisen. Boissier (Voy. Bot. pp. 12-13) recognized that in this species and its allies or varieties the lateral petals vary indefinitely in form, and his D. peregrin uvi covered species and varieties with them either orbicular and subcordate, or elliptical and more or less narrowed below. DeCandolle (Fl. Fr. vi. p. 641) appears to have been the first to describe a species, D. junceum DC, having elliptical lateral petals, his D. cardiopetalum and D. gracile having them suborbicular. In addition to the form of the petals, the aggregate species varies indefinitely in leaf-cutting and habit, the racemes being sometimes dense and compact, and at others elongate and lax, while the leaves are firm or flaccid, close-set or distant, so that a subdivision into fixed species has not proved satisfactory. Perez Lara (Fl. Gad. p. 89) observed that D. longipcs Moris (Z). peregrinum var. longipes Boiss.) always has suborbicular lateral petals, and proposed the following arrangement : — D. peregrinum Linn, a genuinum (D. peregrinmn Boiss.). Lateral petals elliptical, attenuate at base, the racemes either short or elongate [D. junceum DC). (i cardiopetalum (Z). cardiopetalum DC). Lateral petals sub- orbicular, truncate at base. Sub var. longipes {D, longipes Moris). Lateral petals orbicular, subcordate. Subvar. gracile [D. gracile DC). Lateral petals ovate, cordate. This arrangement has the defect of restricting typical D. ijcre- grinum Linn, to a form with elliptical narrowed lateral petals, which its author did not intend, also of making no substantial difference between subvarieties longipes and gracile, which seem to me identical. A better arrangement, which I am following in my Flora of Gibraltar, is that of Boissier (I. c), though the definition of his aggregate species and in part the synonymy are my own, viz. : — D. peregrinum Linn. Lateral petals suborbicular, truncate or subcordate at base, or elliptical and more or less narrowed below. Var. confertum Boiss. (Z). cardiopetakmi DC, D. halteratum Sibth. & Sm.). Racemes dense. Var. longipes Boiss. (D. longipes Moris, D. junceum DC, D. gracile DC). Racemes elongate, lax. In the neighbourhood of Gibraltar I have only seen the form of var. longip)es with truncate or subcordate lateral petals, which is very common in all sandy places, but the following are recorded : — D. gracile DC Uncultivated sandy places and cornfields at Algeciras, Beverchon. D. peregrinum Linn. var. confertum Boiss. Near Algeciras, Schott. Var. longipes Boiss. Catalan Bay and on the Neutral Ground, Boiss., Kelaart, Dautes, and on the slopes of San Roque, Boiss. 12 THE JOURNAL OF BOTANY Umbilicus citrinus, sp. nov. (^ CotyleI. {U. penduUnus DC. var. bracteosus Willk. in Willk. & Lange, Prodr. Fl. Hisp. Suppl. p. 213.) Herl)a caule circa 0-5 m., foliis raclicalibus peltatis, longe petiolatis, caulinis ; racemis secundis, circa 15 dec. longis, floribus horizontalibus vel pendulis, pedicellis 2-5-3 mm., bracteis hyalinis, pedicellis duplo longioribus, calicis lobis triangulari- bus acutis, vix 2 mm. longis, corolla flava, 9-12 mm. longa, cylindrica, sub lobis contracta, lobis ovato-lanceolatis, nervo valido excurrente notatis, carpellis corollas tubo duplo breviori- bus, antherarum basin vix attingentibus, antheris filamentis longioribus. Differs both from U. korizontalis DC. and U. penduUnus DC. in being taller, in its secund racemes of longer bright yellow flowers, the corolla-tube constricted below the longer broader lobes, and in its much shorter carpels. It grows sparingly by the Almoraima Soto (No. 2127), and in a neighbouring valley near Long Stables. Eeverchon's No. 571 from shady woods near Grazalema also belongs here. Umbilicus pendulinus DC. var. truncatus, var. nov. Herba foliis omnibus basi truncatis vel subcordatis, profundius lobatis, petiolis lateralibus. The cauline leaves of this species are frequently laterally petioled, but in the variety even the lowest radical have that peculiarity. On roofs of houses in Palmones Village (No. 751). Sbdum Winkleei, comb. nov. (S. liirsutum subsp. hceticum Kouy. Umbilicus Winkleri Willk.) Specimens sent to Willkomm by Winkler, from the S. Carbonera were described by him as Umbilicus Winkleri in Act. Soc. Bot. Germ. 1883, p. 268, and figured in 111. Fl. Hisp. i. pi. 74 a. Later, in 1887, Eouy, in Bull. Soc. Bot. Fr. xxxiv. p. 441, described a subspecies bceticum of Sedum liirsutum All., based upon specimens sent from the S. de Palma by Eeverchon, differing from the type in being larger, more glandular, leaves longer, and flowers twice the size. I have seen no specimens from either collector, but one collected by Porta and Rigo on the S. Carbonera, labelled Umbi- licus Winkleri Willk., and plants seen there, as well as on the Alcadeza Crags and in the S. de Palma, agree exactly with Willkomm's figure and description, as well as that of Eouy, and I have little doubt that both authors have described the same species. Eouy, however, has overlooked the important feature of the plant — that the petals are united at the base for about one- third of their length, instead of at the base only — which doubtless caused Willkomm to place it in Umbilicus, but as this character is variable in Sedum, and the plant so closely resembles a large variety of S. liirsutum, I propose to keep it in the genus under the name given above. It grows in considerable quantity on rather damp rocks on the summit of the S. Carbonera (No. 1293), in the Waterfall Valley in the S. de Palma, and on the x\lcadeza Crags. Specimens GIBRALTAR PLANTS 13 by Bourgeau from S. de Monchique, Algarve (No. 1872), and S. de Guadarrama above Cbozas (No. 2218), also belong here. Calluna vulgaris Salisb. var. depressa, var. nov. Planta prostrata, raniis tortuosis, dense intertextis ; racemis paucilioris. Plain east of Queen of Spain's Chair, near Gibraltar, abun- dantly (No. 48). Its habit gives this variety a very distinct appearance, and no other form occurs in the immediate neighbourliood, though the type is abundant on the mountain itself and in other parts of the district. Euphorbia gibraltarica N. E. Br., sp. nov. (§ Anisophyllum). Herba annua, caulibus prostratis, tenuibus, a basi plerumque divari- cate ramosis, ramis 8-12 dcm. longis, basi undique longe sed parce hirsutis, apicem versus superne brevius puberulis, inferne glabris, foliis oppositis, breviter petiolatis, 12-20 mm. longis, 6-9 mm. latis, oblongis, basi obhque cordatis, apice rotundatis obtusisve, sgepe macula lata fusca notatis, argute nee profunde serratis, serraturis apiculatis, inferioribus longe sed inconspicue pilosis, superioribus glabris aut basi longe ciliatis, stipulis minutis, libris, triangularibus, circa 1 mm. longis, involucris in inflorescentite furcibus solitariis et per 3 terminalibus, anguste obconicis, in pedicello eis aequante aut superante sensim attenuatis, glabris, glandulis 4, minutis, appendicibus suborbicularibus aut transverse ellipticis, 0-5 mm. latis, intensius roseis ; capsulis 2 mm., globosis, Isevibus, glabris ; seminibus 1-4 mm. longis, 0-8 mm. latis, nigre- scentibus, subtetragonis, angulis tribus carina pallidiore obtusa notatis, quaterno anguste canaliculate, foveolato-reticulato. On railway ballast at Second Venta, near San Koque (No. 2192). Near E. Peplis Linn., but differs greatly in seed and other characters, and quite distinct from any European Euphorbia. Asphodelus serotinus, sp. nov. Herba rhizomae fibris elon- gatis, carnosis, fusiformibus ; foliis ut in A. microcarpo, at Sfepius glaucissimis ; caule lj-2 m., vulgo 1'5 m., graciliusculo, pallido, glauco, saepe ramosissimo ; pedicellis nee sub anthesi nee statu fructifero multo patentibus, paulo supra basim articulatis ; floribus eis A. microcarpi similibus ; periantho 15 mm. longo, albo, vitta olivea-viride lineato ; bracteis pedicelli articulum inferiorem cequan- tibus, late lanceolatis, pallidis, vitta centrali fusca ; staminibus perianthium aequantibus vel paullo longioribus, stylo multo brevi- oribus, lilamentis ad medium usque minute puberulis, ungue 3 mm. longo, latitudine quam longitudine paullo majore, valde ciliato, faciebus subglabris ; capsula quam in A. microcarpo minore, 6 mm. lato, 4^ mm. longo, valde pyriforme aut in collum pedicelli apice hemisphaerico angustiorem contracto. Anthesis Aprili serotino, quo tempore A. microcarpus defloratus est. Near A. microcarpiis Viv. but differs in time of flowering (late April), altitude, and especially in the form of the capsule. Dried specimens look very like small-fruited forms of ^4. micro- carpus, but there is no doulDt that it is quite distinct from any form of that species growing in the neighbourhood of Gibraltar. 14 THE JOURNAL OF BOTANY As I did not discover it till A. inicrocarjms was quite past flowering, I was unable to compare the claws of the filaments in the fi'esh state, but from my notes they are broader in proportion to their length than those of A. microcaiyus. Abundantly from the Alcadeza Crags to Boca de Leon, in the Cork Woods, and succeeds A. viicrocarpus in that neighbourhood (Nos. 1818, 1961). Eynchospora glauca Vahl var. pauciseta Turrill, var. nov. A planta typica foliis angustioribus, spiculis majoribus, setis paucioribus minoribus recedit ; a var. chinense C. B. CI. setis paucioribus minoribus distinguitur. Neither typical B. glauca Vahl nor any variety of it has been recorded before from Europe. The present plant is therefore extremely interesting, both as extending the known distribution of the species and also as forming the type of a new variety. The original B. glauca was described by Vahl from specimens collected in Jamaica, and the present plant differs from it in having long, narrowly linear leaves (a somewhat variable character in this species), spikelets up to 6 mm. long, and either no bristles, or from one to three (most often two) much reduced ones, which are always shorter than the nut (excluding the beak), and sometimes only represented by small protuberances less than 1 mm. long at the base of the nut. The Gibraltar plant comes much nearer the variety chinensis C. B. CI., and the only constant distinguishing features are found in the reduced bristles of the former. If the variety chinensis is kept as a species, as Boeckcler, its first describer intended, the variety pauciseta should be considered a variety of it rather than as a variety of B. glauca. At the Waterfall, Garganta del Aquila, S. de Palma, Algeciras, Wolley-Docl, Nos. 1348, 2088. Atropis iberica, sp. nov. Caespitosa, culmis 6 dec, intensius glaucis, foliis juventate conduplicatis, tandem subplanis, 2-5-3 mm. latis explanatis, nervis utraque facie 3-4 subprominentibus, ligula ad 5 mm. longa, ovali-lanceolata, ex basi decurrente acute acumi- nata, paniculis 16-24 cm. longis, laxis, paulo nutantibus, deorsum visu SEepius subunilateraliter triangularibus, rarius symmetricali- bus, ramis ex quoque nodo 2-3, ineequalibus, patentibus, longi- oribus basi breviter nudis, pedicellis plerumque 1-2 mm. longis, spiculis variegatis, 8-10 mm. longis, flosculis 5-6 (rarius ad 8)5 mm. longis, glumis conspicue inaequalibus, explanatis ovato-lanceolatis, acutis, inferiore uninervia, superiore trinervia quam inferiore duplo longiore, ad mediam usque flosculi proximi attingente, paleis suboequalibus, inferiore acuta, vel breviter acuminata, edenticu- lata, subvalidius quinquenervia, nervis exterioribus subprominen- tibus, basim versus saepe pubescente, superiore dense ciliata, antheris 3 mm. longis. Ab A. Foucaudii Hack, differt ligula elongata acuta, gluma inferiore 1-, superiore 3-nervia (baud 3- et 5-nerviis), magis acutis, paleis acutis vel acute acuminatis, apice anguste membranaceo- marginatis integris, antheris duplo majoribus. ALFRED RUSSEL WALLACE 15 In considerable quantity in the sandy bed of the Palmones Eiver, near Algeciras, within tidal influence (2062). Dr. Stapf, who has very kindly revised my description, informs me that there are Portuguese specimens at Kew, labelled A./estucceforniis. ALFEED RUSSEL WALLACE. (1823-1913.) By the death of Alfred Russel Wallace, which took place at Broadstone, near Wimborne, on November 7th, the last of the giants of English nineteeth-century science is removed. He was born, of Scottish ancestry, at Usk, Monmouthshire, January 8th, 1823; "educated" at Hertford Grammar School, which he left before he was fourteen ; and apprenticed to an elder brother who was a land-surveyor. This employment was distasteful : his attention had already been turned towards Natural History, and, as in so many other cases, Humboldt's Personal Narrative had fired him with a desire to visit the Tropics. It is noteworthy that he began by collecting British plants, though he was eagerly reading books of travel, so that when, during a short time in 1844-5, when he was acting as a master in the Collegiate School at Leicester, he made the acquaintance of Bates, then already an ardent entomologist, it required but little encouragement to make him decide to start for America. He himself says {Travels on the Ainazon, Pi-eface): — "My attention was directed to Para and the Amazon by Mr. Edwards's little book, A Voyage ujj the Amazon, and I decided upon going tliere, both on account of its easiness of access and the little that was known of it compared with most other parts of South America. I proposed to pay my expenses by making collections in Natural History, and I have been enabled to do so." Writing to Bates at the time, he expressly says that one of their objects must be the collection of facts " towards solving the problem of the origin of species " ; but, although they were not then published, it must be re- membered that Darwin had then not only received the initial suggestion of the theory of natural selection from reading Malthus on Population in 1838, but had, in June, 1842, and during the summer of 1844, written out the first and second abstracts of his theory. Wallace and Bates sailed for Para in April, 1848 ; and a year and a half later they were joined at Santarem by Spruce, another Collegiate School master, who, encouraged by Bentham and Hooker, and probably also, as Wallace suggests (Spruce, Notes of a Botanist on the Amazon and Andes, Introduction, p. xxxiii.), by what he heard from entomological friends at the British Museum of how successful Bates and Wallace had already been, had determined to undertake the botanical exploration of the region. The three collectors separated, Wallace first ascending the Eio Negro to the Uaup6s. In September, 1851, Spruce writes 16 THE JOURNAL OF BOTANY from ManAos (then Barra do Rio Negro) to John Smith, the Curator at Kew, that Wallace had just come down from the frontier bringing sketches of several palms, many probably new. Three months later he informs the same correspondent that Wallace, who had started up the Eio Negro a month before Spruce had done so, was " almost at the point of death from a mahgnant fever," whilst his younger brother, Herbert Wallace, who had come out with Spruce, had succumbed in the previous May. Wallace, however, having fortunately sent home his first two years' collections, started for England at the end of July, 1852. The vessel in which he sailed took fire, and the bulk of the specimens he had with him, his sketches and notes, were destroyed. After drifting ten days in open boats, Wallace and the crew were picked up; but the voyage had lasted eighty-two days when he landed in England on October 18th, 1852. In 1853 Wallace published his little book on the Palms of the Amazon, illustrated from his own sketches. Though useful at the time, it was practically superseded by Spruce's classical " Palmse Amazonicae " in the Linnean Society's Journal, vol. xi. (1870). The same year saw the publication of Travels on the Amazon and Bio Negro, Bates's Naturalist on the Amazons appearing in 1863, and Spruce's Notes of a Botanist on the Amazon and Andes (edited by Wallace) not till 1908. Wallace's journal abounds in botanical notes, and contains one brilliant chapter specially devoted to the vegetation of the Amazon Valley. Few passages in his writings are better known than the para- graphs in this chapter in which he contrasts the gloomy solemnity of the tropical forest with the brilliant colours of temperate land- scapes. In 1854 Wallace started once more for the Tippies, reaching Singapore in July, spending in all eight and a half years in the Malay Archipelago, and collecting in Sumatra, Java, Timor, Celebes, Borneo, and New Guinea. An essay, written at Sarawak in February, 1855, and published in the Annals and Magazine of Natural History for September, 1855, " On the Law which has regulated the introduction of new species," is even more impor- tant in the history of biogeography than in that of biogenesis. Though it attracted the attention of Lyell, Darwin, and Huxley, Wallace was disappointed to find that it obtained little general recognition. It was after reading Malthus's book, as Darwin had done just twenty years before, that Wallace, while prostrated with fever at Ternate in February, 1858, wrote the essay " On the tendency of varieties to depart indefinitely from the original type," which he sent to Darwin, and which was read, together with Darwin's chapter " On the variation of organic beings in a state of nature," on the momentous July 1st, 1858, at the Linnean Society. Everyone is familiar to-day with the story of the admirable magnanimity with which the two great naturalists recognized each other's work. " I have felt all my life, and I still feel," writes Wallace in 1870, " the most sincere satisfaction that Mr. Darwin had been at ALFRED RUSSEL WALLACE 17 work long before me, and that it was not left for me to attempt to write The Origin of Species. I have long since measured my own strength, and know well that it would be quite unequal to that task." On the other hand, Darwin writes to Wallace: — " You are the only man I ever heard of who persistently does himself an injustice, and never demands justice." In March, 1859, Wallace wrote to Dr. Sclater from Batchian accepting, with some suggested minor alterations, the six zoolo- gical provinces that Sclater has proposed ; whilst another essay, written about the same time, " On the Zoological Geography of the Malay Archipelago," gives further details as to the boundary between the Indian and Austsi'-alian regions that he located ^ in Lombok Channel. Circumstances thus forced upon his attention the problems of the geographical distribution of animals, and both as collector and as writer he became a zoologist rather than a botanist. The sale of his Malay collections brought him a small fortune, which, when invested, yielded a modest income for a single man ; but in 1866 he married the daughter of William Mitten, the bryologist, by whom he had a son and a daughter ; and his subse- quent hfe in England was one of unremitting literary toil, at first in London and later at several successive country homes. The two fascinating descriptive volumes on The Malay Archipelago, published in 1869, were followed in 1876 by his magnum opus, the classical Geographical Distribution of Animals, which he himself described as an endeavour to do for the twelfth and thirteenth chapters of the Origin of Species what Darwin's own Animals and Plants under Domestication had done for the first chapter. Island Life, first published in 1880 and enlarged in the second edition of 1895, was supplementary to the main treatise, and had appended to it an elaborate treatment of the two subsidiary questions of the Glacial Period and the permanence of continents and ocean-basins. In this work there is a considerable amount of botanical matter. Profoundly influenced by the briUiant suggestions of Edward Forbes, Wallace was always impressed by the importance of geological history in dealing with the past and present distribution of land and water. He made much use of such considerations in modifying Croll's theory of the Glacial Period ; and, though considered by a younger antagonistic school the champion of the permanence of continents and oceans, he constantly accepted very extensive distributional interchanges of land and water. The World of Life, one of his last works, deals with new evidence on the same questions. In 1881 Wallace was granted a Civil List Pension : in 1882 the University of Dublin honoured itself by conferring upon him the degree of LL.D. ; and other universities followed suit at later dates. From his receipt of the Royal Medal of the Eoyal Society in 1868 to the award of the first Darwin- Wallace Medal by the Linnean Society in 1908, Wallace's manifold services to biology have been fully recognized by his confreres ; and he was naturally one of the earliest recipients of King Edward's Order of Merit. Journal of Botany. — Vol. 52. [January, 1914.] c 18 THE JOURNAL OF BOTANY This is not the place to deal with his many labours in various other fields, such as psychical research and land nationalization, nor can we do more than mention the valuable volume of essays 0)1 Natural Selection issued in 1875, and the popular exposition of the whole theory of evolution, as he understood it, in Darivinism (1889). While he differed from Darwin in his views as to the ap- plication of the theory to man, Wallace constantly asserted — even more strongly than Darwin himself had ever done — the sufficiency and controlling effect of natural selection, as opposed to the various post-Darwinian views on evolution. G. S. BOULGER. SHOBT NOTES. Hypericum Desetangsii (Journ. Bot. 1913, 317). — Dr. A. Thellung, of Zurich, has kindly drawn my attention to three important papers upon the above plant and its nearest allies, viz., A. Frohlich, in Sitzungsberichte d. k. Akad. d. Wiss. Wien, math.- natw. Kl. cxx, 505, 1911, and the same writer in CEsterr. bot. Zeitschr. Ixiii, 13, 1913. A. Thellung in Allgem. Bot. Zeitschr. xviii, 18, 1912. In the last-named paper H. Desetangsii is reported from "England"! I wrote to Dr. Thellung for further details, and he replies : — " The Hypericum from England seen by me was found by Prof. Dr. Hans Schinz, in the summer of 1903, in Camborne, Cornwall, growing spontaneously in the garden of the Eev. Hooper. The specimens are to be found in the botanical museum of Zurich University." I am glad to be able to add West Lancashire, v.-c. 60, to the list of counties possessing H. Desetangsii, as Mr. J. A. Wheldon recently sent me an example collected by him, labelled — " H. dubium forma. Bank of the Lune near Caton, W. Lancashire, Aug. 1900," which is undoubtedly the same as the Lewes plant. — C. E. Salmon. Cumberland and Durham Plants. — In 1911 and 1912 my friend Mr. A. Wallis sent me examples of the following plants from the above counties : — Durham, v.-c. 66. Ornitliopus perpusillus L. In considerable quantity in one or two places on the Seaton sandhills. 1911. Interesting from the fact that Tate & Baker remark (Fl. Northumb. & Durham, 1868, 152), "Not seen anywhere recently." — Centaurium pulchellum Druce. Seaton sandhills. 1912. Not reported before, I believe, for Durham. — Chenopodium glaucum L. Slag heap. Old Hartlepool. 1912. Known there for twelve years. — Polygonum litorale Link. Tees estuary and Seaton sandhills. 1912. I think those who separate P. Baii and maritimum as species should, to be consistent, equally keep apart litorale and aviculare. — EupJiorhia Esula L. On sandhills form.ed over an embankment of slag at Teesmouth. 1911. Cumberland, v.-c. 70 (1911). Cerastium tetrandrum Curt. Sandhills near Drigg. — Euphrasia Eostkoviana Hayne. iNear Stye Head Farm, — E. scotica Wellst. Borrowdale. This may SHORf NOTES 19 be an aJdifcion for the county. — Statice humilis C. E. S. f. nana C. E. S. Esk estuary, near Eskmeals Station. — Polygonum Baii Bab. Esk estuary. — Juncus Geranli Lois. Esk estuary, Drigg. — Utricularia ocliroleuca Hartm. Ennerdale Lake. — C. E. Salmon. Note on Symphytum. — There has, I think, been much con- fusion between Symiiliytum perecjrinum of Ledebour and S. asper- 7'ivium of Bieberstein. I suppose this is partly from tbeir close resemblance when they are not in flower. They both grow in my neighbourhood (Tunbridge Wells), and I have not seen any men- tion of a difference I find in the shape of the petioles of their root -leaves. A cross-section of these will show this clearly. Li S. peregrinwn the proportions of this are 4J (wide) by 4 (antero- posteriorly) ; in S. asperrimum 3^ (wide) by 7^, the groove on its upper surface being much deeper and narrower. This distinc- tion is quite lost in the dried and pressed specimen, but in the living plant always available. One of the plants has a much wider limb, of a paler blue than the other. This I take to be the true S. asperrimum. If in this I should chance to be wrong, the distinction will still hold good, though in the reverse direction. A corresponding section in the case of S. officinale purpureum has its lower side (dorsum) much more widely curved than either of the others. — Edward G. Gilbert. Plants of Scilly.— While sojourning among the Isles of Scilly in September last, I landed on the Great Ganinick — a conical pile of granitic rocks matted together by a dense growth of bracken, brambles, sea-beet and grasses — and found on the top Calamagrostis Epigeios in considerable quantity. This, I believe, has not been previously noticed in Scilly, and is a rare plant on the Cornish mainland. In a marsh near the coast on St. Mary's a peculiar form of Juncus maritimus was abundant — so plentiful that, a few weeks later, the crop was mown and carted away to thatch a cottage roof. In its ordinary state /. maritimus is short and stiff, with a panicle that is far exceeded by the sharp-pointed lower bract. The Scillonian plant is weak and tall, 4 to 5 ft., and the panicle is mostly larger and more diffuse, with a lower bract that never exceeds it, and is often not more than a sixth or a quarter its length. The only variety I find described — /. rigidus Desf. — does not fit my plant. I suggest, therefore, that this should be known as var. atlanticus. Of Eupliorhia Peplis, which could not be found in Scilly some sixty years ago when Ealfs searched for it, we saw fifteen plants. — Jas. W. White. REVIEWS. Mikrochemie cler Pflanzen. By Hans Molisch. Pp. ix. and 395, with 116 text-figures. Jena : Fischer. 1913. Paper, 13 marks ; cloth, 14 marks. The importance of microchemical tests in the study of the anatomy and physiology of plants has long been recognized, though it has been greatly exaggerated by some writers and under- 20 THE JOURNAL OP BOTANY estimated by others. At first glance, nothing could appear simpler or more obvious than the proposition that, since the reactions by which any given substance is recognized macrochemically will be yielded by that substance when the test is made under the micro- scope, the different kinds of cell-walls and cell-contents may be demonstrated by the use of reagents which either impart charac- teristic colours to walls and contents, or act as selective solvents, dissolving some of the walls and contents and leaving others un- dissolved, or produce precipitates whose nature furnishes evidence regarding the character of the substance that has united with the reagent to produce the precipitate. As is well knowm, valuable results have been obtained in chemistry and geology by the application of the microscope to the examination of small quan- tities of solid and liquid substances, but when we are dealing with the cells and tissues of plants, considerable difficulties are pre- sented. The microchemical examination of a drop of water, even when several substances are dissolved therein, is a simple matter as compared with that of a plant cell containing perhaps a hundred different chemical compounds, and among these various colloidal bodies which interfere with crystallization and other reactions. On the other hand, it must be remembered that in many cases a microchemical test may afford the only practicable method for the detection of substances which are present in quantities too small for macrochemical analyses, and that microchemical methods, used with due precautions, have many striking additional ad- vantages. The pros and cons of the subject are, however, admirably discussed by Molisch in the introduction to the book under review, which will be found of the utmost service to students of every branch of pure and applied botanical science. The author states that he has been engaged in the preparation of this book for more than twenty years, and that practically every reaction described has been repeatedly tested by him, the result being that the work stands in a class quite apart from the numerous com- pilations devoted partly or entirely to vegetable microchemistry which have hitherto been published. Throughout his descrip- tions of the modes of occurrence and methods of detection of the various substances present in the cell-walls and wall-contents, the author emphasises the need for caution in the interpretation of results and for obtaining confirmatory reactions in those cases where at present we have no reliable and certain method of demonstrating the presence of a given substance — e.g., various glucosides and alkaloids — in the cell, or where the reaction is with probability or certainty to be ascribed to post-mortem chemical change, and so on. The author might with reason and advantage have pointed out still more explicitly that, while many of the results obtained by macrochemical analysis of plant extracts are vitiated by the neglect of investigators to distinguish between substances actually present in plants and those formed in the processes of extraction and testing, the necessity for caution in the interpretation of results is infinitely greater in the case of THE SIMPLES AND DRUGS OF INDIA 21 microchemical tests, the apparent simplicity of which tempts the unwary to rash conclusions ; and that in a large degree it is the uncritical use of microchemical methods that has led to the some- what sweeping and unjust condemnation which these methods have received from various quarters. The author gives a concise account of methods, with a list of reagents, in the somewhat brief general portion of the book (pp. 1-36). The remainder (special portion) is divided into four sections, dealing respectively with inorganic bodies, organic bodies, the cell- wall, and inclusions of the protoplast and cell-sap. Full references to literature are appended to each section, and a good index facilitates the use of the book, which is illustrated by numerous excellent figures, the great majority of these being original. The work will be of the greatest value to teachers wishing to plan a course of instruction in microchemistry, while the lists of plants in which the various substances described occur, render the book of special interest to workers in systematic anatomy, since there can be no doubt that microchemical charac- ters frequently give reliable indications of affinity or otherwise. F. C. Colloquies on the Simples and Drugs of India. By Garcia da Orta. New edition (Lisbon, 1895). Edited and annotated by the Conde de Ficalho. Translated, with an introduc- tion and index, by Sir Clements Maekham, K.C.B., F.E.S. 4to. Pp. xxi, 509. London: Henry Sotheran & Co. 1913. This is the first English version of a very rare book, the third book printed in India in 1563, and " full of printers' errors." The work became known by Clusius's epitome of it, Antwerp, 1567, where the author is given as " ab Horto," a translation of his name into Latin ; it further underwent a change into " Del Huerto," and is thus catalogued by the careful Dryander in his Banksian Library Catalogue. The original Portuguese text does not seem to have been reprinted adequately until Count Ficalho did so, as mentioned in the title-page of the volume now before us. (There was a faulty reprint issued in 1872.) We have therefore to thank Sir C. Markham for translating the entire work from the Portuguese into English, providing an introduction and notes, and, with the help of Sir George Bird- wood, giving the modern equivalents of the Indian names of the plants discussed, the addition of Acosta's figures, with three indexes, and copious notes. The introduction states that Garcia da Orta was born in or about 1490, at Elvas, near the Spanish frontier. He reached Goa in 1534, where he is believed to have ended his days about 1570, having been practising as a physician in India for thirty-six years. It must have been several years before this that he was persuaded to put upon recoi'd his great knowledge of Indian drugs, of which this is the result. It is drawn up in a series of fifty-nine colloquies or conversations between da Orta and his friend Dr. Euano, " the man in the street," a recent arrival at Goa, well read in the old 22 THE JOURNAL OP BOTANY authorities such as Dioscorides and Phny, and ready to quote from them, to which opinions da Orta frequently opposes his own experience. The first colloquy is concerned with the meeting of the two speakers; after that, the subject-matter of each discourse is usually confined to one drug, as Aloes, Amber, Camphor, Cinnamon, till in the last " Betel and some other things, in which some mistakes throughout the work are amended, which have been left through forgetfulness," as it is naively expressed. In 1578 Cristobal Acosta, a medical man of Burgos, published his Tractado de las drogas medicinas de las Indias orientales, Burgos, 4to ; also turned into Latin by Clusius in 1582. The chief source of Acosta's text was the work of da Orta, but he added figures, and these_ cuts, twenty-three in number, have been reproduced in Sir Clement Markham's translation. The determi- nations of the plants mentioned are due to Sir George Birdwood, as already mentioned, to whom the volume is dedicated. It closes with indexes of persons and authorities quoted, of names of drugs, and finally of names of places. Many subjects are discoursed of in a simple and entertaining manner, but an attempt to instance any would take us too far, and exceed the space at our disposal. For any one who has a taste for botanical archaeology or the history of pharmacy, the present volume will be a very welcome addition to his bookshelves. The translator deserves the thanks of such for putting at our disposal an English version of one of the rarest volumes on Indian drugs. B. D. J. BOOK-NOTES, NEWS, &c. Messrs. A. Brown & Sons, of Hull, London, and York, announce for publication " an entirely new work bringing the vegetational history of the county quite up-to-date " — The Vege- tation of Yorkshire — by Mr. F. Arnold Lees, which will be issued to subscribers at 12s. Qd. net. The Preface, which is subjoined to the circular announcing publication, is so characteristic of the " free popular style which has always marked the author" that we venture to reprint it : — " Prepared originally, from personal observations and printed records stretching over 25 years, for the Yorkshire Naturalists' Union (who commissioned it), the inability of that body to issue it — the reasons do not concern the present venture — enables the Author and Publisher to unite to give this important work — on the lines of the Botanical Survey, using the three variously up-to-date Floras (Baker's, Lees's, and Kobinson's) as a foundation —a worthier format, and a much wider dissemina- tion than would otherwise be possible. It is no exaggeration to say that in the botanical world this ' Greater Flora ' (J. D. Hooker) has been looked forward to, and its tardy completion urged on, in many wide apart circles, not only in England and Europe, but from New York to the Antipodes. Its subject — essentially an analysis of the wild vegetation of England's largest BOOK-NOTES, NEWS, ETC. 23 and most vari-surface county, is one that has bearings upon and is apphcable in its factual incidences to many of the still larger island areas of the Temperate zones. No source of science delving into the facts of the past has been neglected ; the question of ' fossil ' seeds in earthy deposits, ancient or more recent, as well as sea-bed dredgings, etc., etc., which might throw light upon the origins, and persistings or passings of its floral features from century to century, has been systematically ' gone into,' and made the basis of a classification which departs widely from the worn, useless one of ' Natives,' ' Colonists ' and ' Aliens ' (for all plants must, at some beginning have been the last at first, even if the first at last in a usefully tentative view), while the autho- rities for all assimilated lemmas of nationale will be found fully acknowledged. The diction of a pen guided by a mind trained to insight as regards the question at issue, combined with a ' free,' ' popular ' style, such as has always marked the author, will, it is believed, make the present work as distinct a step in advance as the Flora of 1888 was over those of the Babingtonian Days to which it succeeded." M. C. Howard has issued a third volume of his important work, Les Zoocecidies des Plantes d'Eitrope, including those of both shores of the Mediterranean (Hermann, Paris, price 10 francs), which consists of a supplement representing the work on the subject during 1909-12. No student of galls can afford to be without this careful compilation, which contains numerous illus- trations. There is an excellent bibliographical index, as well as one of the plants mentioned as hosts. The chief interest, from the point of view of this Journal, of Mr. Aubyn Trevor-Battye's handsome and attractive volume con- taining accounts of his Gamjying in Crete (Witherby, 10s. 6d. net) centres in the appendix devoted to a consideration of the Cretan Flora, of which he gives a general view. The characteristic features of the flora are enumerated, with special notes upon some of the more interesting species, e. g. Acer creticum, of which the leaf-modifications by environment are described. The notes on the disposition of the forests, which are formed by Gupressus sempervirens var. horizontalis, Quercus Ilex, Pinus halepensis and P. Laricio are interesting : the author was much struck by the mischief caused by forest fires, which destroy the seedling trees, and made representations concerning this to the Government Department concerned with the forests of the island, which it is hoped may be attended with success. A list of the more con- spicuous plants of the island and a brief bibliography bring the botany of the volume to a close. It is to be regretted that the proofs were not more carefully read : the names of plants are frequently misspelt, nor is the carelessness confined to them — e. g. Sibthorp's dates are given as i 573-91. The narrative which forms the chief part of the volume is interestingly written, and is enlivened with numerous illustrations. Mr. W. H. Johnson has compiled for use in connection with the study of the principles of agriculture in West African schools 24 THE JOURNAL OF BOTANY a little volume on Elementary Tropical Agriculture (Crosby Lock- wood, 3s. 6d. net), which will it is hoped also prove useful in other tropical countries. It deals simply and clearly with the various parts of a plant, with chapters on soil, food, fungoid diseases and insect pests ; a second part is concerned with the school garden and various matters connected with cultivation. There are twenty useful illustrations, and the book is admirably printed. Mr. A. Bruce Jackson has printed for private distribution, at the request of the Duke of Northumberland, A Catalogue of Hardy Trees and Shrubs groiving at Alhury Park, Surrey — a companion volume to that on the trees of Syon House, published in 1910 and noticed in our volume for that year (p. 296). The book is divided into two parts, one dealing with the " gardens," the other with the " woods " ; this necessitates a certain amount of repeti- tion, and we should have thought the two might well have been combined, indicating by a prefixed initial which plants were found only in one or other of the divisions. The list has evidently been done with much care, and is very nicely printed ; we note very few slips — "Phillyr(^a" (p. 25) — is one. There are brief but useful notes ; references to important works are given ; and a short history of the estate is given as a " foreword." Dr. Henry Franklin Parsons, who died at Croydon, Surrey, on the 14th October, was an excellent " all-round " botanist. Born at Frome, Somerset, in 1846, of a large family, more than one member of which are well known in the world of art, he graduated in medicine with distinction in the University of London. His assistance in the Flora of his native county is acknowledged by its author and a manuscript list of plants observed by him is preserved in the Taunton Museum. Appointed Medical Officer at Goole, he paid considerable attention to the Yorkshire Flora, contributing a paper on maritime plants to the Naturalist in 1875, drawing up botanical reports for the Natura- lists' Union for 1877 and 1878, and assisting Dr. Lees with his Flora (1888). Becoming Medical Inspector to the Local Govern- ment Board in 1879, a post which he held for thirteen years when he became Assistant Medical Officer, he devoted much of his leisure to the Croydon Natural History and Scientific Society, of which he became a Vice-President. In addition to geological and meteorological papers he contributed to its Transactions papers on the times of flowering of spring flowers (1897), on the flora of the commons near Croydon (1899, 1902), and on London casual plants (1906), and gave educational addresses to sections of the Society on hepatics, lichens, and other topics. He conducted many of their excursions and fungus forays, drawing up careful reports of the species observed, and also communicated an annual report on the weather in its relation to local vegetation down to 1912. He retired from the Local Government Board in 1911. For many years Dr. Parsons had also acted as examiner in sani- tary science for the University of Cambridge. His herbarium has been bequeathed to the Somersetshire Archaeological and Natural History Society, and is placed in the Taunton Museum. 25 PLANTS FEOM THE EKET DISTRICT, S. NIGERIA COLLECTED BY Mk. AND MrS. P. AmAUEY TaLBOT. (Plate 529.) (Concluded from p. 9.) Apocynace.e. Gabunia Dorotheae Wernhcam, sp. nov. Omnino glaber- rimus, ramulis complanatis striatis fistulosis ; foliis subcoriaceis late oblanceolatis v. elliptico-oblongis, basi acutis vix acuminatis, venis secundariis conspicuis plus minus distantibus, petiolo brevissimo v. obsolete, stipidis valde conspicuis cavernulas altas axillares necnon vaginas interpetiolares formantibus ; conjmhis ad 15-20-floris, pedunculis longis folia nee tamen superantibus, basi ramulis adnatis, pedicellis longiusculis ; cahjcis segmentis ovatis obtusis ; corolla tubo basin versus contorto insuper gracili, lobis oblanceolatis. No. 3387. Leaves 20-26 cm. x 9-11 cm., with stalk at most 2-3 mm. long ; secondary nerves 8-14 pairs. Peduncles Q-11 cm.. Pedicels 1-1'5 cm. Calyx 4-5 mm. long ; corolla-tnhe 4-5 cm., lobes 1-1-1-4 cm. long, and to 4 mm. broad. Anthers sessile, 1 cm. above the base of the corolla-tube. Nearly related to G. eglandulosa ; it differs chiefly in the relatively shorter lobes and longer tube of the corolla. Voacanga eketensis Wernham, sp. nov. Frutex ramuHs complanatis glabris demum pallidis fistulosis ; foliis papyraceis ellipticis utrinque angustatis obtusiusculis, supra glabris subtus in venis prominentibus minute flavo-puberulis, petiolo brevi puberulo supra altissime canaliculato basi ampliato ; inflores- centiis dichotome umbellatis, in axillis summis gcminis, omnino glabris, bracteis caducissimis, pedunculo s^pe decurvato glaber- rimo ; calycis late campanulati tubo intus supra basin a glandulorum zona induto, lobos latos rotundos demum reflexes superante ; corollce tubo supra contorto, in medio et sub lobos constricto, intus sparsissime hie inde puberulo, lobis patentibus oblanceolatis basi anguste oblongis, tubum superantibus. No. 3388. Leaves 13-16 cm. x 6-7'5 cm., secondary nerves 10-12 pairs, petiole about 6 mm. Peduncle 4-6 cm. Calyx-txxhe 6-5 mm., the internal gland-zone about 2-5 mm. above the base; lobes 4'5 mm. x 5*5 mm. Corolla-ixxhe 1 cm., lobes 1-1-2 cm. Anthers 4 mm., including the slender straight tail 2 mm. long. Near V. picberiila K. Sch., but much more glabrous, the inflorescence being quite so ; the proportionate lengths of tube and lobes in both calyx and corolla differ in the two species, and the anthers are much smaller. Voacanga glaberrima Wernham, sp. nov. Frutex omnino glaber, ramulis teretibus novellis valde complanatis et sulcatis ; Journal of Botany. — Vol. 52. [February, 1914.] d 26 THE JOURNAL OF BOTANY foliis magnis ellipticis utrinque angustatis subacutis petiole basi inflate supra alte canaliculate; inflorescentia dichotoma pauciflora laxa, longepedunculata, due quemque ad nedum in axillis eppesitis, bracteis caducissimis ; fiorihus mediocris, pedicellis longiusculis gracilibus, calyce infundibulari majuscule, tube lebos eblonges ebtusissimes paullum superante, intus glandu- lerum linea paullulum supra basin indute ; corolla tube infra medium apicemque versus censtricto, aliquante centorte, lobes erecte-patentes eblonges obtuses subaequante ; staminihiis omnino inclusis. No. 3389. A very distinct species, with nearest affinity the Liberian V. caudiflora Stapf, from which it is readily distinguished by the large bread leaves and bread corolla-lobes. The leaves measure 18-29 cm. X 8-12 cm., with stalk net much mere than 1 cm. long at most; secondary nerves 14-18 pairs. I'eduncles 11-5 cm., each dichotomizing into secondary peduncles 2-5 cm. long ; ultimate umbels with 2-3 flowers ; 'pedicels 6 mm. ; calyx-ivSiQ 7-8 mm., lobes 6 mm. ; co?-o//a-tube and lobes each 1-1-2 cm., the latter up to 3 mm. broad. Pleioceras glaberrima Wernham, sp. nev. Frutex scandens glaber ramulis novellis gracillimis cemplanatis, demum certice rugose rubro indutis ; foliis lanceolatis ad anguste ellipticis, utrinque angustatis, acuminatis, utrinque in venis ipsis subtus conspicuis glabris, petiolatis, subcoriaceis ; cymis trichotemis minute et obscure pubescentibus pedunculatis in ramulorum summorum furcis singulis orientibus, bracteis minutis ; calycis segmentis evatis obtusis ; corollce extus glabrae latiuscule cam- panulatae lobis oblongis obtusis tubum multo superantibus, lobe quoque supra basin squama oblonga carnosa obtusa apice mode libero, hujus etiam utroque latere in basi ipso appendice filiformi apice 4-fido capitate inserta onusto ; antheris caudis valde in- curvatis. Main read from Oron to Eket, 28 miles, mostly in farm- clearings. No. 3038. Similar to the previous species, its nearest ally, but distinct in the climbing habit, the single axillary inflorescences, and the shape and venation of the quite glabrous leaves. The latter measure 6-5-11 cm. x 2-4*3 cm., with stalk 5-8 mm. long ; secondary nerves 6-8 pairs. Inflorescence 15 cm. wide or more, with peduncle up to 3-5 cm. Calyx 3 mm., corolla-tnhe 3 mm., lobes 7 mm. ; filiform appendages 5 mm. Anthers 2-2-5 mm. " Corolla-lobe cream-yellow shading toward flower-centre with brownish-red. Corolla-scales mauve-pink ; filiform appendages each with 4 bright-yellow knobs " (Mrs. Talbot). Pleioceras Talbotii Wernham, sp. nev. Frutex scandens glaberrimus latice viscose, ramulis Itevibus ; foliis lanceolatis ad anguste evalibus acuminatissimis acutis, brevissime petiolatis ; paniculis densiusculis, j9e(Z/ce//iS siepe longiusculis tenuibus, calycis lobis rotundatis, corollce lobis ovate-lanceelatis subacuminatis acutis, appendiculis filamentesis 10 simplicibus nee capitatis. PLANTS PROM THE EKET DISTRICT, S. NIGERIA 27 Oron-Eket main road, in farm-bush. No. 3008. Distinct in its habit, its glabrous branches and acuminate leaves, acute corolla-lobes, and the simple filamentous appendages one on each side of each stamen. Leaves up to 7 cm. x 2 cm., acuminate for ^ to |- their length, stalks not exceeding 3-4 mm. Pedicels to 5 mm. or longer. Calyx barely 1"5 mm. Coro//a-tube 4 mm., lobes nearly 5 mm. long. " The latex is used as bird-lime. Calyx palest green, shaded purple at base, corolla-tube mauve, limb cream-yellow, mauve shade in centre of flower. Appendages very fine yellow " (Mrs. Talbot). Pleioceras oblonga Wernham, sp. nov. Frutex 3-pedalis ramulis laevibus striatis ; foliis plerumque elliptico-oblongis ad oblongis, breviter acuminatis acutis, basi obtusis v. subrotundis breviter petiolatis, utrinque glaberrimis; paniculis laxiusculis sub- corymbosis ; calycis lobis subrotundis ; corollce tubo lobos ovatos obtusos multo excedente, appendiculis filamentosis simplicibus necnon conspicue capitatis. In drift-sand, or in the bush. No. 3111. Eelated to the preceding species ; diiiers especially in habit, and the much smaller corolla-limb with broadly ovate lobes, as well as in the capitate appendages. Leaves to 10 cm. x 4 cm., acuminate for 1-2 cm. at most, petiole 4-5 mm. Peduncle 3-4 cm. Pedicels to 8 mm. long. Calyx 1-5 mm. at most in length. Corolla-ixuhQ 3-4 mm., lobes barely 2 mm. long. " Flowers cream-yellow with ring of red in centre of corolla- limb " (Mrs. Talbot). Pleioceras Stapfiana Wernham, sp. nov. Frutex ramulis glabris cortice rugoso fusco-rubro indutis ; foliis pergamaceis ellipticis basi obtusis v. subrotundatis, breviter acuminatis obtusis, utrinque nisi subtus in venis obscure et interrupte minute puberulis, venis secundariis saepius plus minus obscuris utrinque 5-7, petiolo brevi sparse pubescente; florihus in cymis terminalibus 2-3-chotomis ramulis divaricatis novellis sinuosis minute pubes- centibus complanatis laxis diffusis dispositis, bracteis plurimis brevibus ovato-lanceolatis subobtusis ciliatis, pedicellis calycem sgepius superantibus; calycis segmentis ovatis breviter acuminatis obtusis ; corollce extus insuper minute pubescentis latiuscule campanulatse lobis oblanceolato-oblongis tubum multo superanti- bus, lobo quoque supra basin squama oblonga obtusissima apice modo libero, hujus etiam utroque latere in basi ipsa appendice filiformi apice trifida inserta onusto ; antJieris caudis incurvatis. Along the rivers. No. 3390. Leaves 9-12-5 cm. x 4-6 cm. ; petiole 6-9 mm. Peduncle Oto 4 cm. Inflorescence attaining 16 cm. in width. Calyx 3 mm. long. Co?-o//a-tube 3 mm., lobes 7"5 mm. ; filiform appendages nearly 5 mm. long. Anthers rather more than 2 mm. Related to P. Gilletii, differing especially in the ample lax D 2 28 THE JOURNAL OF BOTANY diffuse inflorescences with long peduncles. I have named the species in deference to Dr. Stapf who has been unsparing with his assistance in my examination of the Apocynaceae. An interesting member of the collection is a species of Stapf's genus Cyclocotyla (Kew. Bull. 1908, p. 259) allied to Crasiyido- spermum, of which there is ample material, including three or four mature flowers. The genus was based on a plant collected by Pynaert in the Congo Free State, and named C. congolensis. I have examined the type in the Kew Herbarium ; this bears no open flowers. The leaves with their close transverse venation and the curiously cylindrical corolla-tubes are strikingly similar in C. congolensis and in Talbot's plant. On dissection, however, I find that each of the two ovary-chambers contains but two ovules placed one above the other — not five or six in two rows as in the Congo species. Further, in Talbot's plant, the disc is very deeply lobed, and the " calyx-cup " to which the genus owes its name {vide Stapf, loc. cit. 260) is not noticeable ; and finally the inflorescences are axillary as well as terminal. The expanded corolla-limb is extremely small, and after the fall of the corolla the calyx closes rapidly over the top of the ovary. The ripe fruits would be interesting, but unfortunately none are available. The characters just named led me at first to conclude that Talbot's plant should be referred to a new genus allied to Cyclocotyla. But the principal differences between the two may be due merely to immaturity in the specimens of G. congolensis. So that in the absence of further material it will be well, perhaps, to regard our plant as a second species of Cyclocotyla for the present, differing from the prior species in the closer leaf -venation, the axillary inflorescences, the shape of the disc, and the number and arrangement of the ovules. Cyclocotyla oligosperma Wernham, sp. nov. Frutex scan- dens glaberrimus, ramulis gracilibus ; foliis oppositis pergamaceis ellipticis v. oblongis breviter acuminatis obtusis basi acutis petiolo gracili longiusculo, venis secundariis plurimis approximatis trans- versis ; cymis paucifloris laxiusculis alaribus v. terminalibus, pedunculatis. Pedicellis calycem multo excedentibus. Calycis tubo lobos subaequante. Corollce tubo validiusculo limbum angus- tum multo superante, lobis subrotundis brevibus. Ovarii integri quoque in loculo ovula 2. Kwa Ibo E. No. 3052. Fl. through the dry season. Leaves about 10 cm. x 4 cm., petiole about 1"3 cm. ; lateral nerves, thirty pairs or more. Peduncle up to 2 cm., or longer; pedicel to 5 mm. Calyx 3-5 mm. deep, about half occupied each by lobes and tube. C'oroZ/a-tube 8-5 mm. long ; limb 5'75 mm. in diameter. AntJiers nearly 5 mm. long, sessile at about 2 mm. above the base of the corolla tube. Style about 1 mm. long, of which about one-third or more is occupied by the stigma. " Calyx and peduncles palest green. Corolla-tube cream, limb white with primrose centre. Stem and petioles shaded bronze ; leaves very dark green " (Mrs. Talbot). PLANTS FROM THE EKET DISTRICT, S. NIGERIA 29 ASCLEPIADACE^. Tacazzea 2^cdicellata K. Schum. var. occidentalis N. E. Br. Fine specimens of this rare plant, represented hitherto only by the type-specimen at Kew from Lagos. No. 3265. Tylophora smilacina S. Moore, sp. nov. Gaule volubili gracili folioso puberulo hac atque iliac ramuloso ; foliis petiolatis cordatis apicem versus cuspidatis apice obtuse acutis optime 5-nervibus papyraceis utrinque (costis pag. inf. microscopice pube- rulis exemptis) glabris ; cymis solitariis interaxillaribus sparsim puberulis 2-3-cymulosis cymulis sat distantibus subumbellatis saepius 2-5-floris ; bracteis minutis pubescentibus ; pedicellis gracilibus calyce multo longioribus puberulis ; calycis lobis ovato- oblongis obtusis extus pubescentibus ; corollcB mediocris rotatae lobis ovatis obtusis glabris ; corona phyllis tuberculosis late ovoideis antherarum basin attingentibus columnae stamineae basi pauUulum amplificat^ omnino adnatis ; antheris erectis breviter appendiculatis ; stigmate incrassato convexo antheras paullo ex- cedente. Near the Sacred Lake, Ikotobo, in thick bush. No. 3252. Leaves 4-5-6 x 2"5-3'5cm., drying brownish green above, paler below, very shining, nervation conspicuous especially the five main nerves ; petioles usually 1-1-5 cm. long, slender, puberulous. In- florescences mostly 6-8 cm. long. Cymules up to 1*5 x 1"5 cm., but usually smaller, the two lower frequently l'5-2 cm. apart. Pedicels 7-10 mm. long. Calyx divided almost to the base, rather more than 1 mm. in length. Corolla pale crimson, when moistened nearly 1 cm. across, apparently purple, the lobes 4 X 2-5 mm. Corona pale, 1 mm. long. Pollinia minute. Stigma reaching 1 mm. beyond the tips of the anthers, 1 mm. in diameter. To be inserted next T. conspicua N. E. Br., which has leaves diffei'ent in several respects, considerably larger flowers with broader corolla-lobes, the staminal column greatly widened below, larger dark purple coronal-tubercles and a stigma which, while shorter than the anthers, is depressed at the apex and has a small convex boss in the centre of the depression. T. liherica N. E. Br. A second Nigerian locality for this, the other being Oban. No. 3360. Ceropegia Talbotii S. Moore. A second locality — the type- specimens from Oban. No. 3357. LOGANIACEiE. Strychnos (§ Intermedin) eketensis S. Moore, sp. nov. Verisimiliter frutex scandens inermis ecirrosusque ; ramulis folio- sis, aliquanto tetragonis longitrorsum late sulcatis cortice glabro laete brunneo nitente circumdatis ;/o/iis breviter petiolatis oblongo- ovatis apice saepe breviter cuspidatis ipso obtusis obtusissimisve nonnunquam emarginatis basi obtusis vel rotundatis crasse cori- aceis glabris utrinsecus sed praesertim pag. sup. nitidis costarum lateralium jugo basali margin! approximato mox anfracto difficili- 30 THE JOURNAL OP BOTANY usque aspectabili jugo alfcero prominentiore ad 6-8 mm. supra basin costas central! imposito dimidio foliorum abaxiali marginem leviter appropinquante arcuatoque costis et costulis (ut reticulum laxum) subtus magis eminentibus ; cymis foliis plane brevioribus axillaribus pedunculatis satis patentibus bracteatis bracteis late subulatis rigidis ; floribus breviter pedicellatis pentameris ; calycis segmentis suborbicularibus ci'assiusculis inferue regulatis micro- scopice ciliolatis ; corollm triente inf. indivisae tubo calyci a3qui- longo faucibus villoso lobis triangulari-oblongis obtusiusculis crassiusculis in sicco dorso rugulatis ; antheris basi villosis ; ovario subgloboso sparsim piloso quam stylus inferne incrassatus paullo breviore. No. 3237. Leaves 13-17 cm. long, 6-8 cm. broad, in the dry state usually greyish green on the upper and brownish on the lower side ; mid- rib very prominent below ; petioles stout, channelled above, quite glabrous, 5-7 mm. long. Cymes usually 5-8 x 4-6 cm., the peduncles and branches puberulous, with very minute hairs ; bracts 2-3 mm. long, shortly sheathing at the base, the inner face of which bears a number of glands. Pedicels usually 1-2 mm. long. Buds ovoid. Calyx nearly 2 mm. long, the lobes up to 1"5 X 2 mm. Corolla with tube 2 mm. in length and lobes 4 X 1'5 mm. Filaments 1-25 mm. long ; anthers broadly oblong, obtuse, 1 mm. long. Ovary 1"25 x 1 mm. ; style 2 mm. long. Very near S. mcmccyloides S. Moore from Oban, but certainly different, on account of the glabrous branches, broader leaves, longer, open cymes, differently shaped buds and much larger flowers. Gaertnera eketensis Wernham, sp. nov. Frutex (?) glaber ramulis gracilibus ; foliis pergamaceis lanceolatis v. ellipticis v. oblongis acuminatis subacutis, basi angustatis breviter petio- latis, venis secundariis paucis distantibus reticulo interveniente plus minus transverso, stiimlis membranaceis vaginam plus minus persistentem uno latere fissam formantibus apice irre- gulariter setosis ; i)anicula terminali laxa thyrsoidea, ramulis minute pubescentibus, bracteis linearibus ; floribus pentameris brevissime pedicellatis ; calyce brevissime dentato v. subintegro ; corollcB extus subglabrte tubo lobos lanceolatos superante, ore villoso ; anthcrarum apicibus ut stilus stigmatibus minutissime puberulis exsertis. No. 3391. Near G. Dinklagei K. Sch., from which it differs, among other points, in the pentaraerous condition of its flowers, with their relatively shorter corolla-lobes and almost entirely included stamens and exserted style. Leaves 10-12 cm. x 3-4 cm. ; lateral veins 5-6 pairs, petioles 5-7 mm. Stijnilar sheath at first about 5 mm. long, with settB about the same length, later increasing to 8-10 mm. Liflorescence about 7 cm. x 5-5 cm. Calyx rather more than 7 mm. Corolla- tube 3'5 mm., lobes 2-5-3 mm. PLANTS FROM THE EKET DISTRICT, S. NIGERIA 31 BlGNONIACE^. Kigelia Spragueana Wernham, sp. nov. Aibor 80-peclalis, foliis oppositis v. ternatis nonnunquam alternatis, foliolis obovatis V. oblongo-ellipticis brevissime acuminatis acutis integris glabris membranaceis utrinque tenuiter reticulatis, petiolulis brevissimis ; alabastris in apice minute apiculatis ; calyce subuequaliter dentato nee manifesto bilabiato dentibus latis brevibus subacutis ; intus eglanduloso ; corollce tubo lobos suborbiculares minuscules multo superante ; disco nee lobato ; ovario glabro, stigmatis lobis ovatis obtusis. Near Mkpokk. No. 3392. Leaflets 11-13, 8-11-5 cm. x 3-5-5 cm., with stalks 3-5 mm. ; lateral veins 6-8 pairs. Calyx 1-9 cm. long, lobes 4-5 mm. Gorolla-tuhe 5-5-6 cm. long, constricted 1-8-2 cm. above the base ; lobes about 1-5 cm. in diameter. Stamens inserted 1-5 cm. above the base of the corolla-tube ; anthers 8 mm. long. Ovary nearly 1 cm. long. " Flowers very dark terra-cotta" (Mrs. Talbot). Near K. elliptica Sprague, with which Mr Sprague has been good enough to make a careful comparison of the present species. They differ, he points out, principally in the subequal toothing of the calyx in the latter ; in K. elliiitica the calyx is strongly bi- labiate. The leaf-arrangement is somewhat curious. In one specimen it is distinctly ternate ; in another the leaves are set both oppo- sitely and alternately. The latter may be due to twisting or adnation. ACANTHACE^. Thunbergia (§ Thunbergiopsis) Talbotiae S. Moore, sp. nov. Caule volubili sat crebro folioso puberulo ; foliis petiolatis ovatis basi late truncatis interdum levissime cordatis apice breviter acuminatis margine distanter dentatis brevissimeve lobulatis raro subintegris basi 5-nervibus firrae membranaceis supra scabridis vel saltem scabriusculis subtus pilis brevibus strigillosis indutis ; florihus majusculis longe pedunculatis ; pedunculis in axillis oppositis solitariis puberulis ; bracteolis ovato-oblongis apice subito brevissime acuminatis nervis paucis parallelis in sicco baud perspicuis humectatisve et reticulum plane aspectabilibus percursis ; calyce 5-angulari subtruncato (brevissime 5-dentato) pubescente ; corolla bracteolas facile superante tubo basi angus- tato superne gradatim amphficato lobis suborbicularibus tubo plane brevioribus; antheris stam. duorum basi bicalcaratis duorum loculo altero calcarato altero mutico ; stigmate infundibulari, subtriangulari. In thick bush near Awa. No. 3391. A slender-stemmed climbing plant. Leaves for the most part 6-5-8 cm. long, and 3-6 cm. broad, often 4-5 cm. in breadth at the very base, where may occasionally bo seen a slight tendency to hastation, acuminate part + 1 cm. long, with a filiform apex ; petioles 7-20 mm. long, excavated above, puberulous. Pedicels 32 THE JOURNAL OF BOTANY 4-5-5-5 cm. long. Bracteoles 35 x 18 mm., thinly membranaceous. Floivers white. Calyx 1 mm. high. Narrowed base of corolla 8 mm. long, 5-7 mm. broad, widened portion 30 mm. long, at the throat 22 mm., diam 8 ; lobes up to 15 x 13 mm. Filaments fleshy, swollen below, 10 mm. long. Anthers shaggy-haired in front, 10 mm. long or somewhat less, the incurved spurs 2-3 mm, long. Disk fleshy, reaching as high as the calyx. Ovary glabrous, 3 mm. long. Style 22 mm. long, glabrous. Stigma 6 mm. in length, at the mouth 4-5 mm., broad, size gradually diminishing till at the bottom it is only 1 mm. in breadth. Capsule not seen. This should be placed in the genus near T. cordata Lindau and T. togoensis Lindau, from both of which it differs in many important respects. Physacanthus inflatus Clarke. Believed to be the first Nigerian record for this Cameroons plant. The flowers are mauve with white centre. No. 3301. Asystasia dryaclum S. Moore. Previously known only from Oban. No. 3114. Justicia Talhotii S. Moore. Previously collected in the Oban district. No. 3396. Dicliptera Talbotii S. Moore, sp. nov. Planta spithamea ramulis sat tenuibus aliquanto anfractuosis angulatis cito glabres- centibus ; foliis parvis ellipticis basi apiceque obtusis tenuiter membranaceis glabris ; involucris 2-floris in glomerulos breves ovatos densos vel densiusculos axillares terminalesve dispositis ; foliis floralibus parvis linearibus subulatis patentibus glabris ; bracteolis exteriorihus oblongo-obovatis acutis 3-nervibus dense albo-incanis ; bracteolis interioribus calycem aequantibus vel paullo superantibus anguste lineari-lanceolatis acutis incanis ; calycis segmentis angustissime lineari-lanceolatis acutis pube- rulis ; corolla, ex bracteolis breviter eminentis tubo cylindrico superne (ut limbus fac. ext.) piloso-puberulo labio postico suborbiculari antico obovato-oblongo 3-dentato, staminibus breviter exsertis. No. 3217. Leaves up to 25 x 13 mm., but usually smaller. Inflorescences at most 2 x 1"5 cm. Floral leaves about 3 mm. long. Larger of the oilier bracteoles 8 x 3*5 mm., smaller 6 x 3-25 mm. ; inner bracteoles 4-5 mm. long. Calyx 4 mm. long. Corolla 9 mm. in length ; tube 6 mm. long, 1-1-25 mm. broad ; front lip bright mauve, 4 x 1'5 mm., hinder white, 3-5 x 4 mm. Filaments 2-5 mm. long ; anthers 1 mm. Ovary 1 mm. long ; style pilose, 4 mm. long. Near D. Elliotii C. B. Clarke, from which it can easily be recognised by the differently shaped hoary bracteoles. Hypoestes Talbotice S. Moore. A second Nigerian locality ; the type is from Oban. No. 3401. Verbenace^. Clerodendron eketense Wernham, sp. nov. Eamulis molUter pubescentibus ; foliis tenuiter membranaceis late ovalibus brevis- PLANTS FROM THE EKET DISTRICT, S. NIGERIA 33 sime acuminatis acutissimis basi rotundatis v. subcordatis supra sparse subtus densius in venis pubescentibus, venis distantibus valde obliquis, petiolo longo dense pubescente, basi indurante persistente ; infloresccntia thyrsoidea laxiuscula omnino dense pubescente, bracteis plurimis parvis lineari-lanceolatis, ijedicellis 8-9 mm. v. longioribus ; calycis ampli minute hispiduli lobis anguste ovatis acuminatis acutissimis tubum late campanulatum superantibus ; corolla tubo pro rata brevi, extus insuper prgesertim densiuscule pilis glandulosis induto, gracillimo insuper plus minus subito ampliato, lobis oblanceolato-oblongis; staminihus porrectis nee longe exsertis. No. 3393. A well-defined species, distinguished by its soft short pubes- cence, the ample thin oval leaves with long stiff stalks directed outwards and downwards, and the large calyces with but a short length of the corolla exserted. The nearest allied species seems to be C. Welivitschii Giirke, but this has the corolla-tube nearly 4 cm. long. Leaves 16 cm. x 10 cm., petiole 3-5-5 cm. Inflorescence about 10 cm. x 6 cm. Calyx l-5-l'6 cm., lobes over 1 cm. Corolla-tuhe 1-7 cm., the lower slender tubular portion 1*3 cm., lobes 4-5 mm. Stamens exserted about 7 mm. APETAL^. By S. Moore. Tylostemon (§ Hexaerhena) confer tus, sp. no v. Frutex ramis ramulisque subteretibus striatis glabris ; foliis saepe majusculis petiolatis ovato-oblongis apice obtusis ipso apiculatis basi cuneatim coartatis tenuiter coriaceis utrinsecus glabris costis lateralibus utrinque circa 10 patentibus marginem versus arcuatis necnon dichotomis ut costa crassa centralis reticulumque sublaxum pag. inf. magis perspicuis ; ixmiculis subsessilibus axillaribus raro terminalibus abbreviatis satis densifloris harum ramis ut pedicelli alabastraque subtilissime fulvo-tomentellis ; bracteis late cymbiformibus extus tomentellis, pedicellis quam flores brevioribus ; perianthii campanulati intus glabri lobis tubo brevioribus suborbicularibus ; staminum serr. 1 et 2 filamentis abbreviatis crassis pubescentibus quam antherae ambitu sub- quadratae brevioribus ; staminodiis ser. 3 columnaribus pubes- centibus basi biglandulosis ; staminodiis ser. 4 minutis, ovatis, pubescentibus ; ovario subgloboso glabro, stylo inferne crasso superne attenuate aequilongo. No. 3399. Leaves when fully grown 16-20 cm. long and 6"5-7'5 cm. broad, othei's ± 10 x 5-5 cm., drying brownish ; petioles stout, terete if not shallowly channelled above, longitudinally wrinkled, usually about 1 cm. long. Panicles many-fiowered, usually not much more than 1 cm. in length, and often less than that, about 1 cm. also in width. Bracts few seen, 1-5-3 mm. long. Pedicels 1 mm. at most in length, usually shorter. Perianth 1-75 mm. long, 1-5 cm. in diameter just before opening; lobes '75 x "8 mm. 34 THE JOURNAL OF BOTANY Fertile stamens -5 mm. long; outer row of staminodes -4 mm. long, the columnar portion a little longer than the diameter of the subglobular glands. Ovary -4 mm. in diameter; style -4 mm. long. Near T. ohscurus Stapf and differing from it chiefly in the congested inflorescences and the shape of the stamens and staminodes. LORANTHACE^. LoRANTHUs (§ Bufescentes) Talbotiorum Sprague in Fl. Trop. Afr. vi. i. p. 1026 (anglice). Alahastris pilis frequenter ramosis dense ferrugineo-tomentosis ; ramis validis (summun 8 mm. diam.) teretibus supra foliorum insertionem ferrugineo-tomen- tosis alibi tandem glabris pallide nitidis lenticellis paucis anguste elliptico-oblongis linearibusve usque ad 6 mm. long, praeditis; foliis brevipetiolatis ovato- vel lanceolato-oblongis sub apice acuto recurvis basi angustissime cordatis 15-17"5 x 6-5-8-5 cm. crasse coriaceis supra glabris nitentibus subtus opacis et costa media furfuraceo-ferruginea exempta glabris ; U7nbeUis axillaribus fasci- culatis pedunculatis 3-4-floris ferrugineo-pubescentibus ; pedun- culis 2'5 mm., pedicellis 1"5 mm. long. ; hractea erecta ovata ellipticave apice rotundata extus pubescente intus glabra 3x2 mm. ; toro calyce cum subcylindrico circa 2 mm. long. ; calyce breviter denudato '5 mm. long. ; corollce extus rufo-tomentellae intus glandulosae tubo fere usque medium unilateraliter fisso (ampulla basali inconspicua ovoideo-oblonga) 6-8 mm. long, basin versus plus minus curvato lobis erectis spathulatis 10-11 mm. long. ; filamentis corollse loborum basi insertis circa 8 mm. long. ; antheris oblongis circa '5 mm. long, harum loculorum septis 4 ; disco minimo ferrugineo-piloso ; stylo glabro sursum metuliformi hujus parte incrassata pallidiori 6 mm. long, collo circa 1 mm. long. ; stigmate capitato 5 mm. diam. Hab. South Nigeria, Oban ; Mr. (C- Mrs. Talbot, 1281. The above is a translation of Mr. Sprague's description. It is inserted here because L. Talbotiorum has again been obtained by Mr. & Mrs. Talbot, this time in the Eket district. No. 3395. L. Braunii Engl. var. Talbotii, var. nov. A typo discrepat prsecipue ob perianthium supra basin parvam inflatam maxime attenuatum, parte attenuata brevissima, in sicco modo 1 mm. long, totidemque diam. Basis inflata 1'5 mm. long. 3 mm. lat. No. 3400. It is very difficult to discriminate between some supposed species allied to L. Braunii, which has led Mr. Sprague to unite several of them with the latter. He was good enough to examine the plant under notice, and it has been named as above in accord- ance with his advice. Euphoebiace^. Grotonogyne Manniana Miill. Arg. Male plant. Previously known only from Fernando Po. No. 3258. C. Zenkeri Pax. Female plant. A Cameroons-Gaboon species, now first recorded as Nigerian. No. 3397. Caperonia latifolia Pax. The leaves are smaller than usual DISTRIBUTION OF CERTAIN BRITISH ALG^ 35 (3-5-4-5 X l'8-2 cm.), and much like those of Sutton Hayes, 703, from Panama. The fii'st Nigerian record for this species. No. 3398. Maprounea vievibranacea Pax & K. Hoffm. Distribution : Cameroons and Spanish Guinea to Belgian Congo ; now first reported as a Nigerian plant. No. 3253. Description of Tab. 529. 1. Branch with inflorescence, natural size, of Talhoticlla cketciisis Bak. fil. 2. Buds with narrow bracteoles, x 2, 3. Flower, x 2. 4. Longitudinal section through flower, x 8. THE DISTEIBUTION OF CBETAIN BRITISH ALG^. By a. D. CottoxV, F.L.S. When discussing the peculiar features of the marine flora of the West of Ireland in the report for the Clare Island Survey (Trans. Roy. Irish Acad. vol. xxxi.part 15), theoretical conclusions with regard to certain species were hampered owing to lack of precise data as to distribution. This applied not only to the Continental range, but also to that in our own country. Three of the most noteworthy were Ptilota pliiiiiosa, Gallitliamnion arlms- cula, and Codium mucronatum var. atlanticum, each of which possesses a somewhat remarkable distribution in the British Isles. All three were plentiful in the warm waters of the Clare Island district, and they occur also on the west of Scotland, but none of them are found in any part of the English Channel or south-east of England. The difficulty with regard to the English east coast admitted of explanation ; but it was not easy to understand why these plants should not extend by way of the Welsh coast to Devon and Cornwall. The algal flora of Wales, from Anglesea to St. David's Head, is but poorly known ; and it was not possible to state in the Report whether the species in question reached this region or whether they stopped short at a point further north. There was also a measure of uncertainty with regard to North Devon and North Cornwall, although this area had been often and thoroughly explored. Unlike some algae, these are conspicuous species and of well defined habitats, so that they could hardly be overlooked if properly searched for. Two trips therefore were made to the Welsh coast during the past season with a view to certifying the presence or absence of these and other species, the districts chosen being Cardigan Bay (Barmouth to Aberystwyth) and Pembroke- shire (Newport Bay to near St. David's Head) ; whilst during a summer holiday at St. Ives some of the north Cornish coast between Padstow and the Land's End was examined. The following notes refer only to the species mentioned above. It appears worth while to publish details of the trips instead of a mere statement of results, as the marine botany of Wales has been somewhat neglected. We may first recall the distribution of the species in question. 36 THE JOURNAL OF BOTANY Gallitliamnion arbuscula is a boreal alga. It is found in Norway, the Faeroes, Scotland and Ireland, and is abundant in the Glare Island district, forming a band on rather steep exposed rocks. In the much colder waters of the North Sea it does not occur south of Yorkshire, and on our west coast it was known to descend to Ayrshire and Isle of Man. Ptilota plwnosa is also a northern species. Found in Iceland, Nova Zemblya, and Spitzbergen, it descends as far as Yorkshire on our east coast, and had been recorded from North Wales and the Isle of Man on the west. On Clare Island it is plentiful, being attached to the stipes of Lami- naria Cloustoni and washed ashore in company with such south- ern plants as Taonia and Gallymenia reniformis. It is supposed to occur right round Ireland. With regard to these two species it was pointed out (1. c. p. 170) that ecological factors appeared more likely to explain their absence on the south and west coasts of Great Britain than unfavourable currents or lack of spores. Both require an open rocky coast, and the scarcity of such ground south of Yorkshire may have limited their range in the North Sea, and prevented their access to the Straits of Dover. But on the west coast this explanation does not hold, rocky ground being plentiful both in north and south Wales. The third species, Godium viucronatum (first correctly identified during the Glare Island Survey), is a much more remarkable plant. Frequent in Australia and New Zealand, and almost identical with the form known as Nova Zelandice, it is in Europe only known from the British Isles. Though long overlooked, it can happily be recog- nised with certainty in the herbarium, and its distribution was carefully worked out for the Glare Island Eeport. It occurs in Scotland from Orkney to Bute, and in the north of Ireland from Antrim to Donegal, and from thence southward to Bantry Bay. On Clare Island it is plentiful and conspicuous. In the Irish Channel it occurs on the Isle of Man, but is absent (as far as is known) from the east of Ireland, from England, and from the rest of Europe ; also from the Atlantic coasts of North America. The plant can hardly have been introduced to Ireland, since it has existed in Bantry Bay for over a century. A special look-out for this alga should be kept, as any data that would throw light on its isolated position in Europe would be highly valuable. 1. Gardigan Bay. — The two regions investigated were those of Barmouth and Aberystwyth, advantage being taken of the British Mycological Society's meeting at Dolgelly in May 1913 to examine the coast-line in the vicinity. The rocks are slaty, Cam- brian and Silurian respectively. In neither locality did the ground appear suitable for the algse sought for. The shore is flat, and does not possess steep, clean, exposed rocks, the well-known habitat for C. arbuscula. No specimens of L. Gloustoni (the usual host of P. i^lumosa) were noted, but it is possible, and indeed likely, that that plant occurs in deeper water. Clean rock- pools, of the type in which Godium viucronatum flourishes, were also absent. At Aberystwyth a large amount of angular gravel is present (probably derived from glacial drift), and the injurious DISTRIBUTION OF CERTAIN BRITISH ALG^ 37 effect of this on the vegetation is very apparent. During westerly gales and high seas the gravel is dashed over rocks ; and even such tough plants as Ascopliijllum are in places shorn off, while the scarcity of many of the more delicate species in pools which appear otherwise suitable, is no doubt to be attributed to the same cause. Through the kindness of Professor R. H. Yapp I was able to look through the algal herbarium at the University College, but though possessing a good variety of species for the type of shore, none of the algae in question were represented. Professor Yapp informed me that similar ground continues as far south as Aberaeron, and that the same type of vegetation obtains: the presence of loose stones and sand would doubtless be specially inimical to Godium mucronatum, and he did not recollect having seen any representatives of the genus in the neighbourhood. The same factor is probably also operative on the flat rocky shore south of Barmouth, though it was not so clearly demonstrated. At Borth the ground is obviously unsuitable. As far as Cardigan Bay is concerned, therefore, we may be fairly safe in saying that over a very large area (probably from Portmadoc to Aberaeron) the three species under consideration are absent, though as shown above this may be largely accounted for by unsuitable conditions. 2. Pembrokeshire (August 18th-21st). — Pembrokeshire ap- peared likely to prove interesting, not only from its rocks being of a different character to those of Cardigan Bay, but from its position at the extreme south-west corner of Wales. It was possible that the rock-loving species in question might obtain a footing in that neighbourhood, and that they might descend as far south as St. David's Head, though they were not known to occur on the coasts of Cornwall, the other side of the British Channel. Fishguard was chosen as a centre ; it is easy of access, and, unlike Tenby and Milford, open to the west. But in this neighbourhood the precipitous cliffs are a great obstacle ; and the time lost in going from one bit of accessible ground to another, when tides waiting for no man are in question, is a serious matter. In a southerly direction the coast-line was examined at irregular intervals from Fishguard to a point about half-way between Strumble Head and St. David's. The cliffs and head- lands are for the most part composed of volcanic rocks, and these descend sheer into the sea ; but bays of slaty rocks, which are more accessible, are also present. The latter are strewn below with stones and gravel, and, as can be seen from above, they support little or no vegetation. Two near Pembrush Point were descended by paths none too safe, but practically no algal growth, save Enteromorpha and a small amount of Fucus, was found. Time did not permit of extending the search to the Head itself, but it is probable that the same type of coast continues. It is very unsuitable for littoral algae owing to the steep cliffs and lack of reefs. North of Fishguard a much better vegetation was met with. 38 THE JOURNAL OP BOTANY Newport Bay, with its wide sweep, was hurriedly searched, and though not possessing a rich flora, was very much better than anything previously seen. On the rocks at the south-west corner (and probably at the north-east also) algal associations of the " exposed " series were found, the band of the short form of Porphyra umhilicalis being well developed. These rocks appeared suitable for Callithamnion arbuscula, but it was absent. Pools sufficiently clean and clear for Codium mucronatum were frequent, but not a plant was noted. Considering the date of the visit (August) the vegetation of this bay showed much variety, and in spring the flora must be a rich one. A certain amount of rocky ground occurs between Goodwick and the old town of Fishguard. These rocks, which are easily reached by boat, possess a fair algal vegetation, but it is of the semi-sheltered type. A large amount of clean drift had collected, the examination of which gave a good idea of the sublittoral flora. No specimens of Ptilota, however, occurred amongst it. By far the best piece of collecting ground w^as a little bay at Dinas. The rocks, slaty but not much broken, are accessible on either side of the bay, and the ground exhibits on the right the exposed type of rocky-shore formation, and on the left the semi-sheltered type, and the sand-and-rock series. These rocks, though they occupy a small area, have a very rich flora ; and the sudden appearance of such a large number of species shows how generally the spores are distributed, and how ready the plants are to thrive when the conditions for their establishment are suitable. For an examination of the flora of the rocky shore of Pembroke- shire this little bay could hardly be beaten ; indeed, for variety of ground and richness of flora in a compressed area, it is one of the best I have met with. By the road, which is hilly, Dinas Bay is four and a-half miles from Fishguard town and six miles from Fishguard Harbour Hotel. In good weather it is a pleasant sail of three or four miles from the G. W. E. quay at Fishguard. On the right side of Dinas Bay the Nemalion association is well developed, and in its upper parts a distinct but rather sparse belt of Gallithamnion is present. It was at once obvious that this belt was composed of C. spongiosum, and not of C. arbuscula, a conclusion which was confirmed by the microscopic examina- tion of a number of specimens both saxicolous and epiphytic. A few pools exist, but Codium mucronatum was hunted for in vain. Ptilota also was not found. This concluded the search in south- west Wales. It is tolerably certain that P. jAnviosa is absent, and that C. arbuscula does not occur. Not a sign of the interest- ing Codium mucronatum was seen on any occasion. 3. North Cornwall (September). — The north Cornish coast is so familiar that it need not be described. Padstow, Newquay, and St. Ives were the spots specially investigated, each of which had been more or less worked by Mr. E. M. Holmes and by the late R. Y. Tellam. St. Ives is poor for algae, but the shore towards the south-west affords here and there some better ground. Both at St. Ives and at Newquay C. arbuscula was altogether absent, DISTRIBUTION OP CERTAIN BRITISH ALG^ 39 though a slight growth of Geranium acanthonotum, which is often a co-dominant species in the association, was present ; Nemalion also, which enjoys similar conditions, formed bands and patclies. At Padstow the ground is too sheltered for the Callithamnion or Nemalion belts, and time was not available for the exploration of the exposed coast-line in that neighbourhood. Pools occur near Padstow, as at Newquay and St. Ives, but Codium imicronatum was completely absent. P. pluniosa also was not observed any- where in Cornwall. 4. North Wales. — As this area had not been visited, an effort was made to obtain information by means of correspondence and examination of herbarium collections. Professor E. W. Phillips, of Bangor, was kind enough to supply notes, and also to lend a number of specimens from the Anglesea neighbourhood (chiefly Menai Straits and Puffin Island) from his private collection. The examination of these showed that whilst P. pluviosa is plentiful in North Wales, the other two plants, Callithamnion arbuscula and Codium mucronatum, are not found. This was in agreement with Professor Harvey Gibson's list of algae for the Liverpool Marine Biological Committee's district (Trans. L.M.B.C. vol. v. 1891, a copy of which was kindly lent me by the author)."^' Professor Harvey Gibsoii writes that he has not found Codium mucronatum on the north coast of Wales, but that Callithamnion arbuscula is plentiful in the Isle of Man, and specimens from that locality exist in the Kew collections. Mr. N. E. Brown, of Kew, has examined the shore around Llandudno, and he tells me he has never observed any species of Codium in the rock pools of that region. We are now in a position to summarize the facts as to the distri- bution of these species. As far as Wales is concerned, P. plmnosa is frequent on the rocky shores of north-west corner, but does not appear to extend south of Anglesea ; Callithamnion arbuscula and Codium mucronatum, on the other hand, though common in the Isle of Man, do not reach the Welsh coasts at all. With regard to the British Isles, the Callithamnion and Ptilota are general in Scotland and descend as far as Yorkshire in the North Sea, and are abundant in the warm waters of the west of Ireland, but in the Irish Channel stop short at the Isle of Man and Anglesea respectively. Both are supposed to occur in the south of Ireland, but this requires confirmation. Codium mucronatum remains as mysterious as ever. No trace of it has been seen in the localities investigated in North or South Wales, and it may be safely stated to be entirely absent from Devon and Cornwall. It is still only known in Europe from the Atlantic shores of Scotland and Ireland (where it has existed since the early part of last century) and from * A not irrelevant addition to that list may, however, be here noted, namely, the ordinary Codium toinentosum, which was found by Professor Phillips and is preserved in his herbarium. A fragment of a frond of Ptilota phuiiosa, in- scribed " Ilfracombe E. T.," also occurs in his collection. But as the plant does not occur in this well-worked region, the locality given must be regarded as erroneous. 40 THE JOURNAL OF BOTANY the Isle of Man. Whether it is spreading at all it is not yet possible to say. Though Cardigan Bay as a whole is not good for marine algae, a certain amount of excellent ground occurs here and there in both the northern and southern parts ; and the presence of a very large number of species at Dinas Bay, Pembrokeshire, indi- cates that algal spores are widely distributed, and that the plants are ready to thrive when the conditions are suitable. Hence it is unlikely that geological factors are wholly responsible for the absence of the algge we have been considering. Their distribution on the south and east of Ireland requires to be reinvestigated, and when this has been done it will be profitable to again con- sider the question of tides and currents. Meanwhile, this con- tribution as to their distribution in Great Britain may not be superfluous. P.S. — Since the above was written, an interesting paper by C. L. Walton on the shore fauna of Cardigan Bay has come to hand (Journ. Mar. Biol. Assn. vol. x. No. 1, Nov. 1913, pp. 102- 113). In describing the geological features of the bay, the author draws attention to the large quantity of residual drift, and its injurious effect on the fauna. He also notes the importance of the dip and strike of the older rocks, and remarks that when the strike is parallel to the coast, if the dip is low the rocks are barren ; if high, with a landward dip, they are also barren ; but if high and seaward, there may be a fairly good fauna on the land- ward slope. With regard to these factors, contrasts as well as similarities will be seen when the algal vegetation is considered. Geological points of this nature have been largely overlooked by writers on algal ecology ; but it is clear that for the more detailed study of the vegetation they must be carefully considered, as they not only directly account for the presence or absence of certain associations, but in the case of hard rocks, especially, largely determine the general contour of the foreshore. The points emphasized by Mr. Walton do not, however, throw any further light on the alg^ discussed above. DOESET PLANTS. By H. W. Pugsley, B.A. During the summer of 1912, and again in 1913, I spent a fortnight at Swanage, which is a good centre for field botany. The following brief notes may be of interest. On both occasions I searched on the east side of Littlesea for Scirjncsparvulus, but with- out success ; and as other botanists have recently similarly failed to find it there, I fear it may have been extirpated by the inroads of the sand, which at some points is obviously advancing towards the lake. In 1913 I was accompanied on several walks by Mr. C. B. Green, who is compiling a very exhaustive list of local records. DORSET PLANTS 41 Fumaria officinalis L. var. Wirtgeni Haussk. — Quite typical near Wareham Eailway Station. Spergularia marginata Kittel. — In the Flora of Bournemouth this is recorded only for Keysworth (near Wareham), but in following the shore of the harbour from Poole to Sandbanks it was found to be nearly as common in the salt-marshes as S. salina Presl. It still occurs on the dry cliffs at Tilly Whim, with Aster Tripoliiwi, as recorded in the Flora of Dorset. Astragalus glycyphyllos L. — A single plant of this was observed in 1912 on a grassy roadside between Swanage and Durlstone Head. Its origin there is not easily explained, as it is not a species likely to be introduced ; and it may occur naturally in some of the enclosed ground in that vicinity. It is recorded for Dorset by Pulteney without locality, and the only specific habitat in the Flora is " Between Ashmore and Kushmore " in the north- east of the county. Atriplex laciniata L. — The Flora of Dorset remarks of this plant " North and South Haven beaches. Bell Salter ; not con- firmed since Pulteney's time," and the Flora of Bournemouth states " an old record not confirmed of late." Mr. Salmon (Journ. Bot. 1911, 365) notes the species for South Haven, and in 1913 it occurred in fair quantity on the beaches at both North and South Haven. Salicornia disarticulata Moss and S. appressa Dum. — Sandy marsh at South Haven. Both confirmed by Dr. Moss. Epipactis latifolia All. — A few plants near Durlstone Head. Hitherto recorded only for Creech Grange, in the Isle of Purbeck. Juncus compressus Jacq. — A dwarf rush seemingly referable to this species grows with /. Gerardi Lois, in a few spots near Littlesea, and some intermediate individuals of possible hybrid origin were also observed. The plant does not match typical /. compressus, as its ripe capsules, while clearly exceeding the perianth-segments, are shorter and more globular ; and it closely resembles a form collected by Mr. J. W. White at Borrow, Somerset, and described in Journ. Bot. 1889, 49. Buchenau reported Mr. White's plant to be intermediate between /. com- pressiis and /. Gerardi, but it apparently finds no place in his monograph in Engler's Botanischen Jahrbiicher, Band xii. p. 185 (1890), or in his subsequent works on Juncus, and would seem to be still an unnamed form. It is intended to deal further with this rush when fresh flowering material can be obtained. J. com- pressus has not been previously recorded for the Isle of Purbeck. Deschampsia setacea Eichter — Wet heath near Stoborough. New to the Isle of Purbeck. MR. JOHN GILBERT BAKER. A PLEASANT function took place at Kew on January 13, when an address of congratulation on the attainment of his eightieth birthday, signed by those who had been associated with him Journal of Botany. — Vol. 52. [February, 1914.] e 42 THE JOURNAL OF BOTANY during his long connection with the Herbarium of the Eoyal Gardens, was presented to Mr. John Gilbert Baker, one of the oldest surviving contributors to this Journal, of which he was at one time assistant editor. The Morning Post of the following day published an account of an interview with Mr. Baker, from which we take the following autobiographical details: — " I come of a family of yeomen farmers who were Quakers, and I was born at Guisborough, in Yorkshire, where my father was a general merchant. My earliest recollections are of the quiet little country town of Thirsk, to whicli my father removed his business when I was only six months old. At the age of nine I was sent to the Quaker school at Ackworth, where I remained for three years, at the end of which time I was transferred to another Quaker school, that of York. Among my schoolfellows at York were Joseph Rowntree, the founder of the well-known cocoa business, John Rowntree, his brother, Henry Seebohm, who became famous as an ethnologist [ornithologist] , and two other brothers, George and Henry Brady, both of whom after- wards became Fellows of the Royal Society, a distinction which was conferred on myself as long ago as the year 1875. The Quaker school at York was a capital place. The discipline was mild, which was not the case, I believe, at most of the schools in my youthful days, and above all special attention was given to the natural science which soon became my delight. Of course, at the time of which I speak, that is to say early in Queen Victoria's reign, scientific study was not nearly so widespread as it is now. My school was the first to institute a Nature Study Society, and to this practically all the boys belonged. We used Babington's Manual, an excellent book of its kind, though costly according to present notions, and we used to go botanising in our leisure time in the fields round about the old Cathedral city. I entered into the pursuit with such enthusiasm that before I had been at the school twelve months I won a prize for the best collection of plants, and was thereupon made curator of our little herbarium. The Headmaster, Mr. John Ford, was not, so far as I am aware, specially devoted to scientific study, but several of the teachers were ardent botanists. I left school at the age of fourteen and went into my father's business, where I remained for eighteen years. During that time I was not wholly engrossed in commercial pursuits. " All my spare time was employed in studying botany, and during this period of my career I wrote my book entitled North Yorkshire : Studies of its Botany, Geology, Climate, and Physical Geography. This was published by Messrs. Longmans, and a second edition was brought out in 1906 by the Yorkshire Naturalists' Union. Three years after this work was published I received a communication which changed the whole current of my life. It was a letter from Sir Joseph Hooker, who had recently been appointed Director of Kew Gardens in succession to his father. Sir William, and in it he offered me the post of First Assistant at the Herbarium under Professor Daniel Oliver. Sir MR. JOHN GILBERT BAKER 43 Joseph was at that time a stranger to me, and his communication came as a complete surprise. I need hardly say how thankfully I accepted his offer. That was the beginning of a long connection with the Gardens that has been to me a source of continual delight. Evenually, as you are probably aware, I succeeded Professor Oliver as Keeper. I had the privilege of assisting Sir Joseph Hooker with some of his scientific books. Sir Joseph was a man of unbounded energy, and, in my opinion, one of the greatest men of science who ever lived. He had been a great traveller, visiting the Antarctic region among other parts of the world in pursuit of his favourite studies. His father. Sir William, was of a more stay-at-home disposition, but he did great things for the Gardens, which, when he first went there, were in a state of absolute chaos. It had been a private garden of the Koyal Family, and in the reigns of George IV. and William IV. had been greatly neglected. Sir William built three great houses, estab- lished communication with all the botanic gardens in the world, brought from Glasgow his herbarium and library, and, in short, made the place for the first time a thoroughly scientific institution. The good work has been well carried on by his successors, Sir Joseph Hooker, Sir W. Thiselton Dyer, and Sir David Prain. During Sir David Prain's comparatively short tenure of office he has built a beautiful series of tanks for hardy water-plants and bog and marsh plants, and has done a lot to the Rockery, while altering the walks in such a way as to give easier access to the temperate house and other points of interest. Sir William Hooker planned out a vast scheme under which all the plants of the British Dominions and dependencies, 50,000 in number, are to be made into a list. That scheme began with Australia, and is not yet finished, although the end is now in sight. The magnitude of the undertaking may be judged from the fact that in India alone there are 13,000 plants, more than are to be found in the whole of Europe. The fact is that in India you have a sort of variation in climatic conditions, from perpetual snow to extreme tropical heat. Africa is still to a large extent an un- explored region from the point of view of the botanist. The flora of the Cape has been completed, and in it are the names of 10,000 plants. In tropical Africa scarcely a day passes but a new plant is discovered. Mrs. Talbot has recently made a collection in Nigeria, which totals 10,000 genera and 200,000 species. Of these 15 of the former are new^ as are 150 of the latter. " The science of botany, I need hardly say, is a very different thing from what it was when I began my studies. Linnaeus is quite out of date, although his system is still useful as a sort of index to plants, but in his day only about 10,000 plants were known in the entire world. The system of Linnaeus is what is called the artificial system, and it was superseded by that of Jussieu, which is the natural system. The difference between the two would be best described by saying that that of Linnaeus was a dictionary in which the different plants are given in alpha- betical order, while that of Jussieu is a grammar in which they E 2 44 THE JOURNAL OF BOTANY are arranged in groups. Jussieu's work was carried on by De Candolle, who classified according to the natural system all the plants known in his time. But when these eminent men wrote, geology can hardly be said to have existed, and our views of botanical science have been profoundly affected by the discovery of the law of evolution. No one now believes that there have been successive creations in the sixteen geological periods. With plants as with animals it has been a case throughout the world's history of the survival of the fittest. The teaching of botany in England was altered under the regime of Sir W. Thiselton Dyer, a great organiser. Before his time lecturers dwelt on the natural orders and external appearance of plants ; now the German method is followed of studying the cells under the microscope and the physiology of plant life. "Anything that would conduce to the welfare of Kew Gardens is of national interest, for it is by far the most important esta- blishment of its kind in the wide world. In the Berlin Gardens they mount their plants on rockeries and group them roughly according to the different mountain systems to which they belong, and this is a very interesting plan. But the Berlin Gardens do not vie in importance with ours, which extend to 300 acres and are visited yearly by a million and a half of people. The existence of such a place for study reflects great credit upon our Government, which has not stinted the means of keeping it up. One of its best friends has been Mr. [Joseph] Chamberlain, who when he was in the Government obtained an extra grant, by means of which the great temperate house, one-eighth of a mile long, which had long remained unfinished, was completed. Mr. Chamberlain has always taken the greatest interest in the Gardens, which he used to visit regularly every year. I well remember those appearances of his, and the orchid which he invariably wore in his buttonhole. As you know, he is a great collector, and we often used to exchange plants with him. His collection of orchids at Highbury, Birmingham, is, I believe, worth £25,000. The great value of certain orchids consists in their rarity. As with other things, it is a question of supply and demand, and prices rise when they are sought after by wealthy collectors. The question of beauty is a subsidiary one. As a matter of fact you can buy some of the most beautiful specimens in existence for five shillings. I remember that we had a rare lily that was going to be photographed, but it was eaten up during the night by a cockroach. That cockroach did not know that his supper cost us something like £10." BIBLIOGKAPHICAL NOTES. LVa. — Miquel's 'Plants Junghuhnian^.' Some MS. notes in the copy of Miquel's unfinished Plantcs JunghuhniancB in the library attached to the National Herbarium supply important corrections to Mr. Dunn's contribution in this BIBLIOGRAPHICAL NOTES 45 Journal for 1913 (p. 358). I have verified these notes, which may be tabulated as follows : — ( [Flora, 1851, p. 302, and Fasc. 1 = pp. 1-106; March, 1851 Wikstrom, .Irs^^era^^. 1851, ( p. 133 (1855).] „ 2= „ 107-270; 1852 [Wikstrom, ^. c] ([Gramineae (p. 341) bears date Febr. 1854 ; see also Wikstr. op. ciL, 1853-54, p. 125 (1856).] [ [Lichenes (p. 427) and Colle- „ 4= „ 395-522; 1857 macese (p. 491) bear date I Febr. 1856.] The enumeration of the Hepatic®, excerpted from vol. v. of Verhandl. dcr K. Akademic Wettcnschcqjpen, 1857, and included in fasc. 4, terminates abruptly. It was evidently intended to issue a completion of the work, for van den Bosch in vol. ix. of the work just cited (" Hymenophyllacege Javanicse," 1861) and in Nederl. Kruidk. Archie/, iv. (" Synopsis Hymenophyllacearum," 1859) quotes pp. 545-571 of " fasc. 5, ined." -^ ^ Wiltsheae. LVI. — "The Department of Botany." The account of the Department of Botany occupies pp. 79-193 of the first volume of The History of the Collections contained in the Nat2iral History Departments of the British Museum which was published in 1904. It is divided into three portions : — 1. General Sketch (pp. 79-84) ; 2. Chronological Account of tho Principal Accessions to the Botanical Collections to the end of 1902 (pp. 85-128) ; 3. Alphabetical List of the more important contributions to the Collection of Plants in the Department of Botany (pp. 129-193). It is stated in the Preface to the volume that the section was undertaken by " Mr. George Murray, assisted by Mr. Britten " : Mr. Murray's name is appended to it at the foot of p. 193 ; and on p. 84 is a note : " Mr. Murray desires to state that advantage has been taken of Mr. Britten's unique knowledge of the history of the botanical collections. He, with Mr. Gepp's help, has completed the work." How far this represents the facts of the case is sufficiently well known to those who were officially employed in the Depart- ment at the time of the preparation of the work ; but it may be worth while to place these facts on permanent record. With the exception of the second section, which was transcribed from the records of accessions by Mr. F. G. Wiltshear, now in charge of the Departmental Library, the whole of the work was done by myself, with the help of Mr. Gepp in regard to the collections of Cryptogams. Mr. Murray's only part in the book was the addition of his autograph and the note quoted above, which was added, though not in the form that we had accepted as satis- factory, in consequence of representations made to him by Mr. Gepp and myself. Mr. Murray did not even read the proofs, although he officially passed them for press. James Britten. 46 THE JOURNAL OF BOTANY SHORT NOTES. EoMULEA CoLUMNJE. — It seems to have escaped the notice of British botanists, who now uniformly adopt it, and indeed of botanists generally, that, on the principle of priority, this name cannot stand. As the Index Ketvensis correctly shows, it is synonymous with Ixia parvifiora of Salisbury, who was the first to give it a specific name. Eichter (PI. Europ. i. 252 (1890) ) cites Salisbury's name in his synonymy of Bomulea Columnce, quoting this latter from its first publication by Sebastiani and Mauri (Fl. Eom. Prodr. 18 (1818) ). Bomulea parviflora of Ecklon, cited in the Index (Top. Verz. 19 (1827) ) is a nomen muhim, doubtfully referred by Mr. Baker (in Dyer, Fl. Cap. vi. 1, 42) to his var. parviflora of B. rosea, and therefore cannot stand. Salisbury's description is from a plant " sponte nascentem in Ins. Jersey, legit E. Finlay." Specimens from this collector, of whom I know nothing more — he was apparently the first to find the plant in the Channel Islands — are in Herb. Banks ; the sheet was indorsed by Dryander: "Guernsey, on the sides of hills in dry places; in great quantity near Irwin's Eedoubt. Jersey, near Grouville, in sandy places, Lieut. Finlay." Other specimens were sent from Jersey by Mr. Gosselin, and it was partly from the latter that the material was received upon which the figure in English Botany was based ; a note by Sowerby on the plate, which is embodied in the printed text, explains the composite nature of the figure — " Flower from a root sent by Sir Joseph Banks, and grown by Mr. Anderson at Kensington ; the remainder from a wild Guernsey specimen sent us dry by Mr. Gosselin." The full synonymy of the plant will be found in Eichter (/. c.) ; in an abridged form it runs : EOMULEA PARVIFLORA, COmb. nov. Ixia parviflora Salisb. Prodr. 34 (1796). Ixia Bulhocodium Sm. E. Bot. 2549 (1813), non aliorum. Bomulea Columns Seb. & Maur. Fl. Eom. Prodr. 8 (1818) et auct. plur. Trichomona Columnce Eeichenb. Fl. Excurs. 83 (1830). It may be noted that both Eichter and the Kew Index erroneously cite the E. Bot. reference as " Trichonema Bulhocodium," and that Ixia Bulhocodium Sm. finds no place in the latter. — James Britten. The Adventitious Flora of a Library Court. — It may be of interest to record the following remarkable list of plants that were growing last summer in the back court of the University Library, Cambridge, all — with one exception — having originated from seeds that had been brought by wind or birds. In one small gravel-covered piece of ground, only nineteen feet square, in the south-west corner, with the buildings rising to a considerable height on two sides, there appeared no fewer than six species of Epilobium {E. angustifolium, E. hirsuUim, E. parviflorum, E. montanum, E. roseum and E. ohscur^tm), besides Tilia vulgaris, Acer Bseudo-platanus, Botentilla reptans, Sambucus nigra, Betula SHORT NOTES 47 alba, Ulmus glabra, Salix caprea, S. caprea x {cinerea ?), S. aurita x caprea, S. {aurita ?) x caprea, Taxus baccata and Asparagus maritimus, whilst Marcliantia polymorpha covered most of the bare places. There were no catkins on the willows, so the Rev. E. F. Linton, who saw them in sitiX, could not with certainty name all of them. The Mulberry {Morus nigra) occurred in this same corner some years ago, but has been removed. On the gravelled border of other parts of the court were to be found Ilex Aquifolium, Gratagus Oxyacantha, Betula alba, Salix caprea, S. cinerea x {aurita ?), and Asplenium Buta-muraria (growing from a joint in the wall). Most interesting were two plants of the sea- coast grass Festuca rottbcellioides, discovered by Mr. Jenkinson, the Librarian, and transplanted by him from Clare College, where numerous traces of it could still be seen in the autumn, though the authorities had done their best to destroy it by means of a weed-killer. Many of the above plants, unfortunately, were ruthlessly destroyed during the holidays by the too energetic gardener. The front court, too, has its interest for botanists. Most Cambridge men have noticed the ferns Pteris aquilina, Lastrea Filix-mas, Asplenium Trichomanes, and the beautiful tufts of Asplenium Buta-muraria, that have for years grown, together with Linaria Cymbalaria, undisturbed, with their roots deep down in the cracks between the Senate House steps — " the best locality for ferns in Cambridgeshire ! " All the ferns excepting A. Tricho- manes occur elsewhere in the court. The list could be lengthened considerably if the names were given of the flowers that make the lawn gay in summer, when it has been left unmown for a short time. In the middle court the most conspicuous plants during the last year or two have been Erigeron canadense (in great abundance), Epilobium montanum, and Linaria Cymbalaria. — G. GOODE. Euphorbia gibraltarica (pp. 10, 13). — The attribution of this name to Mr. N. E. Brown is an error for which I am not responsible. Mr. Brown was good enough to help me in working out the species, but the ultimate naming and description are my own. A. H. WOLLEY-DOD. Utriculakia ochroleuca. — To the list of counties on p. 10 should be added: "92. Aberdeen S. Dr. Trail sp." — Arthur Bennett. BEVIEWS. Biochemio der Pfianzen. By Friedrich Czapek. Erster Band. Pp. xix. and 828. Jena : Fischer. 1913. Paper, 21 marks ; cloth, 25.20 marks. That the first edition of this work, published in 1905, in two volumes, containing 581 and 1025 pages respectively, supplied a generally felt demand is sufficiently indicated by the fact that it has long been out of print. In view of the enormous amount of work published during the last eight years in this extremely pro- 48 THE JOURNAL OF BOTANY gressive branch of investigation, one is at first surprised that the present volume, the general arrangement of which remains much the same as in the first edition, has grown only by about 200 pages. However, a second glance shows that the author has now prepared what is practically a new book rather than a new edition. The explanation, apart from the author's unrivalled power of discrimination and compression, is simple. When Czapek's book first appeared, it was the most extensive work on biochemistry, but in the meantime there have been published numerous treatises dealing with every branch of the subject, including the great encyclopfedia (BiocJiemisches Handlexicon) edited by Abderhalden, Wehmer's Die Pflanzenstoffe, and a host of smaller works, among which we may mention the fine series of Monogi'aphs on Biochemistry edited by Plimmer and Hopkins. It is doubtless owing to the important technical applications of biochemistry that the publication of research papers has been so closely followed and accompanied by that of books summa- rising the results of these researches, with the result that perhaps in no other branch of science can one so readily keep in touch with the latest developments of the subject. Czapek's work, however, stands quite apart from all the others, for it is some- thing more than, and something very different from, a mere ency- clopaedia of the chemistry of plant products. Indispensable though books of the latter class are, there is perhaps even greater need for a philosophical treatise on the chemical physiology of plants, and this is what Czapek has supplied in this new edition of the Biochemie der Pflanzen. Instead of merely expanding the book in order to include the results of work done in biochemistry of plants during the last eight years, Czapek has, by wholesale con- densation and omission of material now available in other compi- lations and in special monographs, been able — without unduly enlarging his work — to give a critical summary of the progress made up to the present time in the direction of ascertaining the nature, relationships, and biological significance of the substances built up and broken down in the course of metabolism. Hence, this still bulky volume, packed with citations and serving as a guide to the extensive literature of the subject, is thoroughly readable, and presents a clear picture of the chemical aspects of plant physiology, which is only rendered the more complete and attractive by the wealth of detail introduced and fitted in place, the author never losing sight of the fundamental principles which the multitudinous details, if less skilfully handled, would tend to obscure rather than to illustrate. The volume opens with a concise historical introduction, in which the author reviews the general progress of phytochemistry from the time of Aristotle onwards. The general section (pp. 20- 239) contains a critical account of the present state of knowledge and opinion concerning colloids and colloidal phenomena, catalysis and enzyme action, immuno reactions, chemical stimulation, &c. — in short, of the whole of the remarkable body of facts and theories resulting from the application of modern physical chemistry to BIOCHEMIK DBR PFLANZEN 49 the problems of vital organization and metabolism. In con- cluding this section, in which the amazing progress of biochemistry is more strikingly reflected than in any other part of the book, the author considers the chemical aspects of heredity, variation, evolution, pointing out that in the light of the recent work of Keeble, Armstrong, and others, various colour varieties and mutants in flowers may be regarded as cases of " chemical mutation," and that many other morphological characters will doubtless be shown by future work to be the outw^ard and visible signs of inner biochemical changes taking place in the cell. The remainder of this volume is devoted to the carbohydrates (in the widest sense) and the lipoids (fats, waxes, phosphatides). It is impossible here to indicate, even by a bare enumeration of the chapter headings, the extraordinarily detailed and compre- hensive nature of the author's treatment of these groups of substances in their various relations to each other and to metabolic processes. Many will doubtless be inclined to consider the amount of condensation to which these chapters have been subjected as somewhat ruthless, so much that appeared in the first edition having been excised altogether in order to make room for new matter based on recent investigations. In many cases, for instance, in the sections dealing with chlorophyll and allied substances, the older work — largely based on impure extracts — has very pi'operly been omitted or relegated in severely condensed form to the valuable historical introductions which preface the main sections throughout the book. In other cases, however, one is inclined to wonder whether the author has not gone rather too far in his fixed determination to condense or omit material dealt with in other text-books, despite the fact that what the work has lost in encyclopasdic fullness (and perhaps dullness) it has cer- tainly gained in other directions. The outstanding feature of the work is the remarkable width of its range, the author having thoroughly ransacked the literature of pure and applied chemistry and botany in his brilliantly successful efforts to sort out and piece together a vast number of scattered observations which thus acquire a significance which would have escaped a writer less critical and less able to estimate relative values, while, on the other hand, a good deal of work which has formed the basis of uncritical and even extravagant theories is here dismissed with scant ceremony. Although the references to literature prior to 1905 have been greatly cut down, this volume contains some 6000 citations. It is almost incredible that any one author can have actually consulted every one of these thousands of books and papers. Hence, it is hardly surprising that one can detect a sHght slip here and there. For instance, in the section deahng with light intensity and its measurement, a footnote reference is given, following the heading " Messende Methodik " (p. 534), to a paper by Wiesner {Flora, Band 105, p. 127), which certainly does not deal with methods of light measurement, though its title (" Ueber die Photometrie von Laubsprossen ") might imply that it does. p ^ 50 THE JOURNAL OP BOTANY Index Keiuensis Plantarum Phanerogaviarum Supplementuyn Quar- tum Nomina et Synonyma ojnnium Generum et Specierum ah initio anni mdccccvi usque ad finem anni mdccccx nonnulla etiam antea edita complectens ductu et consilio D. Prain confecemnt Herharii Horti Regii Botanici Kewensis Cura- tores. Oxonii e Prelo Clarendoniano [Nov.] 1913. 4to, cloth, pp. 1-252. Price £1 16s. Among the many works which issue from the Kew Herbarium, it may be doubted whether any has proved of more universal usefulness to phanerogamic botanists at large than the " Kew Index," to adopt the title by which it is generally known. It is therefore with much satisfaction that we receive the Fourth Supplement, which brings the work down to the end of 1910, and which, considering the arduous nature of the undertaking, has been issued with reasonable promptitude. In a prefatory note Sir David Prain indicates certain altera- tions which differentiate the present from the previous Supple- ments, as well as from the body of the work ; this it may be well to reproduce : — " Supplementum 4'"3i usque ad anni 1910 finem prolatum ita expolire maluimus ut magis botanices studiosis valuisset atque profecisset. Propositae consulto ad eum prsecipue adno- tandae sunt immutationes infra enunciatse. Imprimis annum in quo nomina edita sunt semper designatum est. Iterum nomina antea usitata sub nomina nunc utenda recitata sunt ; nominibus nudis inter synonyma enumeratis nomina accepta addita sunt. Nomina hibridarum arte operatorum negleximus : denique, nomina iam locis setate posterioribus perscripta et ex iis denuo memora- vimus." Small as these alterations are, they are distinctly improve- ments. The omission of dates of publication was, as we have more than once pointed out, one of the few oversights which detracted from the value of the Index and of its Supplements. We are glad, by the way, to note that these are given in the right place, i. c. at the end of the citation ("iv. 348 (1907) " ) instead of in the way frequently employed of late years (" iv. (1907) 348 " ) — a method which possesses no advantages and which, in the case of most periodicals, is positively inaccurate and therefore misleading. It is doubtless convenient that references to readily accessible works should in many cases be added to those indicating place of first publication, but it may be hoped that this will not lead to quota- tions at secondhand, to which it seems somewhat to lend itself. The citation of names already given after those now printed is presented in a new form : these are now preceded by a colon instead of by the sign " =," which in the original work was employed in more than one sense and was thus open to mis- understanding. All names are now printed in the same type ; the use of italics for synonyms is discontinued. The names of garden hybrids, as to which in the Index there was some in- consistency, " some being abolished and others retained," are now entirely excluded : garden names, however, even when nomina INDEX EEWENSIS 51 nuda, seem to be thought deserving of a place : thus we have on the first page " Abutilon album, Ilort. ex Gentile PL Ciclt. Serves Jard. Bot. Brux. 3 (1907) nomen. — Hab.?" — a sufficiently doubt- ful plant. Names omitted from the Index and from earlier Supplements are included ; for a future Supplement it might be well to include a list of such names which will be found in the National Herbarium. The numerous still-born names published by Garsault in 1764 and 1767 are now included — a desirable addition in view of the fact that Gihbert's, many of which are on the same footing, found place in the Index. We note however that Gentaurium majus [Centaurea Gentaurium L.) is omitted — a somewhat important oversight, inasmuch as the association of this with C minus {Erythraa Centaurmm Pers.) has been cited as evidence that Garsault had no claims to botanical knowledge and that his names have no claim to recognition.''' It is of course only by use that the detailed accuracy of such a work can be tested, but so far as a necessarily casual inspection enables us to judge, this Fourth Supplement shows a great im- provement on its predecessors. The new names published in this Journal, which have in former Supplements often been overlooked, seem, so far as we have tested them, to be duly recorded, the only omission we have noticed being, curiously enough, Bazumovia hisiyida, to the omission of which from the Third Supplement we called attention when noticing that volume.! The work is singularly free from misprints — the only one we have noticed is in the third entry (" peudunculata ") — and the references are very carefully done (Berger should replace Eendle as the authority for Aloe pacdogona). We have thus little but praise for this Fourth Supplement : our only regret is that when alterations were being introduced, the insertion of a comma between name and authority should not have been abandoned, in accordance with general custom. And is it too much to hope that Mr. Jackson's long promised introduction to the work may be issued with the Supplement now in preparation ? TJie British Bust Fungi ( Uredinales) : their Biology and Classi- fication. By W. B. Grove. Cambridge University Press. Pp. xii. -{- 412, 290 figs, in text. Price 14s. net. The study of the Uredinales, the group of fungi known more popularly as "rusts," is one of the most fascinating in the whole realm of botany. It is only within comparatively recent times that some understanding has been arrived at with regard to their strange life-histories and their amazing niceties as concerns their particular host plants. One of the first systematic books (cer- tainly the best, if regard be paid to the amount of original work contained therein) to take any account of the biology of the group was Plowright's Monograph of the British Uredinecs and Usti- * See Jouru. Bot. 1909, 322, where in line 5 from top ^' majus" should be " TWjnws." t Journ. Bot. 1908, 267. 52 THE JOURNAL OP BOTANY laginecs, reviewed in this Journal for 1889. Since that time no Enghsh book deahng with the rusts has appeared, and, considering the vast amount of research which has been done on the Continent during the past few years — much of which seems quite unknown to British mycologists — the present book, gathering up this work as it does, should prove a welcome addition to our mycological literature. The first thing to strike one on perusing it is the very small amount of original work from the biological side : an occasional observation such as everyone must make is all that is recorded. The book is divided into two parts : general and systematic. In the former the life-histories of typical Uredinales are described, giving an indication of the variation met with. Other matters considered are sexuality, the nature of the so-called spermatia, the nuclear life-history, alternation of generations, the spore forms and their groupings, specialisation, and immunity. A further chapter deals with classification and phylogeny. As this portion of the book consists of only eighty-four pages, the various accounts are necessarily much condensed. Boom for a little expansion could have been obtained by a slightly different arrangement which would have saved a certain amount of repeti- tion. Most of the recent work on the special points in this section has been considered. It is here naturally that there are differences of opinion. The evidence brought forward of fertilization by " Christman's method " in Puccinia Caricis is quite inadequate. The phylogenetic tree on p. 83, which places the Ascomycetes and the Basidiomycetes on one line of development and the Uredinales and Ustilaginales on another, will not be accepted by many mycologists who understand the questions which arise. The other points in these preliminary chapters call for no comment here save that they seem to have been on the whole fairly stated. The classification adopted differs slightly from that proposed by Dietel in the Appendix to Engler & Prantl's Pflanzenfamilien, followed with minor modifications by the majority of recent authors, but Endophyllum is made the type of a family as in Dietel's first system. The order is divided into the Impedi- cellatee ; Melampsoracege, CronartiaceaB, Coleosporiaceee and En- dophyllacese : and Pedicellatse ; Pucciniaceae. The Melampsoraceee are subdivided into Melampsorese and Hyalopsoreae ; Coleosporiaceae into Coleosporieae, Ochropsoreae, and Zaghouanieae, and Pucciniacese into Puccinieas, Phragmidiese, and Gymnosporangieae, subfamilies proposed at different times by various authors. A generic key is given. In the body of the book the families are taken in the reverse order from that in which they appear in the classification and in the generic key, for which there seems no reason. The author tells us in his preface that the specific descriptions are based upon those of the Monographia Uredinearum of the brothers Sydow. " Those of all the species of which British specimens could be procured have been carefully revised, and there is hardly one of them that has not been added to or amended. Fischer's Uredineen THE BRITISH RUST FUNGI 53 der Schweiz and McAlpine's Busts of Australia have also been found extremely useful " — the former, we should say, particu- larly so. Following the example of recent works line drawings of certain spore forms, usually the teleutospore, are given in most cases. A list of synonyms is given, but there is no attempt at completeness and very few dates are added. It would have been a great advantage to students to have dates in all cases. To confirm dates is a very heavy task, but when an author has really looked up all the references, it gives but little trouble to add the date, and in doing so he greatly helps those who have not access to a large botanical library. Very useful notes are appended to the descrip- tions, giving accounts of recent studies. It is in the gathering together of these scattered facts that the book will prove most useful. Annual critical rSsumes of such work appear in certain journals, but its relation to systematic mycology is often overlooked. The host plant, relative frequency, date, locality and universal distribution are usually all given. Where spore forms have not been observed in this country they are described within square brackets (except in the case of the interesting genus Miles ina). It would seem, however, that we may be accepting very many statements from observers in different countries with insufficient reservation since "as is now known, the life-histories of such heteroecious forms require to be worked out for each country separately." For instance, it seems unsafe to join up the aecidiospores and teleutospores of Gronartium Quercinum described from American specimens when only the uredospores have been found in this country, and " their dimensions are smaller " than in America. The author does not go as far as many with regard to the much named " biological species " and quotes with gusto the rather notorious case where Probst showed that one such species was confined to a form of a variety of a subspecies. It would seem that all the recently proposed genera of the old world have been adopted. There is still some uncertainty as to some of these genera. For example, for the old Ghrysomyxa albida a new genus KUhneola was proposed by Magnus. Into this genus Arthur put Phragmidium Tormentillce. and in this is followed by Mr. Grove. This same species, however, was placed by Magnus himself in the genus Xenodochus and this is accepted by Klebahn. A way out of the difficulty on the present lines would seem to be to make a third monotypic genus. A totally different proposition meets one in the case of the genus Puccinia, where in the present work 137 species are described. As no scientific arrangement of these has yet been proposed, Sydow's method of classifying them according to the host plant is adopted. Certain conceptions of the author as to what must be regarded as species seem rather open to criticism. The suggestion that Triphragmium FiUpendidce should be lumped with T. Ulmaria is one that will not be followed without the proof of inoculation experiments. The nomenclature is according to the International Rules (Brussels Congress, 1910) 54 THE JOURNAL OF BOTANY but in certain cases, e. g. Phragmidmm disciflorum, there seems to have been a shght misunderstanding as to what those Eules entail. A useful index of host plants, one of the species, a list of excluded species, a glossary, and a bibliography — not so complete as one would wish— complete the volume. The name of the publishers is sufficient guarantee of attractive printing and binding. The line drawings (there are also a few wash drawings) are clear. Mr. Grove has compiled a book, reasonable in price, which must be in the hands of all British mycologists. t -n •^ *=■ J. Eamsbottom. Gruppe7itoeise Arthildung. By Hugo de Vries. 8vo, pp. viii. 365, with 121 figs, and 22 coloured plates. Borntraeger, Berlin. 1913. Price 22 Marks. More than ten years ago in his Mutations-Theorie De Vries endeavoured to show that the production of new species was, like any other physiological process, a matter for experiment. Much work has been done on the subject since then by De Vries him- self. Dr. E. Gates, and others, especially in relation to the genus CEnothera, investigations on which played so important a part in De Vries's original memoir. The present remarkably w^ell-illus- trated volume details the results of a large number of experimental crossings between species and forms of CEnothera and the relation of these results to the author's theory of the origin of species by mutation. The text is divided into five sections. The first, entitled "The Origin of Species through Mutation," is an exposition of the author's views on mutations in the light of his theory of intra- cellular pangenesis. The pangens are contained in the nucleus, and have each their special character. Their condition varies, and may be active, inactive, or labile, and the last-named state is the cause of the conditions requisite for mutability. Section ii. deals with " Keciprocal and Double-reciprocal Hybrids." The author points out that the products of reciprocal crosses are very often unlike, indicating that the characters contained in the pollen differ from those in the egg-cell. Such species De Vries terms heterogamous in contrast with isogamous species, in which pollen grain and egg-cell bear the same characters. By the process of gamolysis — that Is, the determination of these special characters by crossing — De Vries seeks to analyse the constitution of numerous natural species as well as mutants. Sections iii. and iv. are respectively headed " Twin-hybrids " and " The Pangenetic Investigation of New Species," and Section v. is a general dis- cussion on the " Causes of Mutation." Appended are a biblio- graphy of the literature of the subject since 1903 and a systematic list of the crossings in the genus CEnothera which are described in the text. The book is an important contribution to the study of genetics ; and not the least useful features are the series of photographic blocks with which the text is illustrated and the well-executed coloured plates at the end of the volume. ABE 55 : BOOK-NOTES, NEWS, dc. Mr. Cbdric Bucknall publishes in the Journal of tlie Linncan Society (xli. No. 284 ; December 29) a " Eevision of Symphytum,'" which, as will be expected by those who read his paper on some hybrids of the genus published in this Journal for 1912 (pp. 332- 337), is evidently a very careful piece of work. He recognises twenty-five species : one of them — discovered by Shuttleworth at Hy6res in 1871, and distributed by him as S. floribundum — is now first described ; other new species are S. anneniacum and S. Bornmuelleri. The species which have been found in Britain as introductions are S. asperum, S. pcregrinum, S. tcmricicm, S. caucasicum, and S. orientate ; the confusion associated with the name asperrimitm, which has been applied to at least seven species, has been unravelled. The history of the genus is fully detailed ; there is a very full synonymy, including that of pre- Linnean authors, and a full list is given under each species of the specimens examined, and — perhaps a little redundantly — the names under which they appear in the herbaria consulted. The descriptions throughout are very full and show much personal investigation : the paper is, in fact, a model monogra.ph, and Mr. Bucknall is to be congratulated on the result of his many years' work. Notwithstanding the plenitude of botanical literature, com- prehensive works of reference on the various branches of the science are all too scarce, and much time is wasted and informa- tion overlooked in consequence. The Bradley Bibliography, which is being issued under the able direction of Prof. C. S. Sargent as Publication No. 3 of the Arnold Arboretum, is a welcome addition to this class of works. It is to be " a guide to the literature of the woody plants of the world, published before the beginning of the twentieth century," complete in five quarto volumes. Two of these, dealing with dendrology, have already appeared ; the first " includes all botanical publications containing references to woody plants, except those which are restricted to a particular family, genus, or species, which are found in the second volume and are arranged according to the system of Bngler & Prantl," and in chronological sequence under each subject. The third volume will be occupied with literature on the economic products and uses of woody plants and with arboriculture; the fourth with forestry, and the last with an index to the whole. Dr. A. Rehder, who has had this work in preparation for upwards of ten years, made a tour of the principal botanical libraries in Europe and America in his endeavour to render the enumeration as thorough as possible. Each title is given at ample length and followed by particulars of size, date and place of publication ; occasional annotations also add to the value of the entries. The volumes are of convenient format, and their fine typography is worthy of the Riverside Press from which they issue. The expense con- nected with the production of this work, which has been made possible by a family gift commemorative of William Lambert Bradley, must have been considerable, and is reflected in its cost, $100, or nearly £21.— F. G. W. 56 THE JOURNAL OP BOTANY A LONG and interesting account of William Gardiner (1808- 1848), contributed by Mr. Alexander P. Stevenson to the recent part (xxvi. pt. 2) of The Transactions of the Botanical Society of Edinburgh, contains a note reprinted from Gardiner's Botanist's Bepository which may throw some light on the difficulties attend- ing the verification of George Don's discoveries : The innkeeper at Auchmithie, hearing Don's name mentioned, " pronounced a warm invective against that gentleman, who, he observed, had ruined these trees, for since he had been prowling about there, not a plant worthy of notice was to be seen. I had no reason to doubt mine host's assertion, for I have frequently searched Mr. Don's habitats in vain. I verily believe his plan respecting rare plants was — first to dig up all the specimens he could see, and then note the locality." The Second Circular of the International Botanical Congress for next year, issued last month, contains the "program of work" for the Congress, which was defined by the Congress of 1910 as follows : — " 1. To fix the starting-point for the nomenclature of {a) Schizomycetes (Bacteria) ; {h) SchizophyceoB (excepting Nosto- caceae ; (c) FlagellataB ; (cZ) Bacillariaceae (Diatomacese). " 2. To compile lists of nomina generica utique conservanda for (a) Schizomycetes ; (b) Algae (incl. Schizophyceae, Flagellatae, &c.) ; new lists for groups not included in the list of 1910 and also a supplementary list ; (c) Fungi ; {d) Lichens ; (e) Bryophyta. "3. Compilation of a double list of nomina generica utiqtie conservanda for the use of palaeobotanists. " 4. Discussion of motions relating to neio points which were not settled by the Eules adopted at Vienna in 1905 and at Brussels in 1910." The English members of the various committees for the consideration of these points are — for Mosses, Mr. Antony Gepp ; for Algae, Mr. Gepp and Mr. A. D. Cotton ; for palaeobotany, Dr. Arber, Dr. D. H. Scott, and Prof. Seward. Copies of the Circular, which defines the functions of the various Committees, may be obtained from the General Secretary, Dr. Eendle, to whom inquiries regarding the Congress may be addressed. Dr. Eendle is also a member of the Editorial Committee, " functioning as a Permanent Bureau of Nomenclature." The new edition of the Biographical Index of British and Irish Botanists is pi'actically ready for press, and it is hoped that it may be printed before the end of the year. The attention of our readers may be called to the list of little-known British botanists ■which was printed in this Journal for 1912 (pp. 61, 130, 194) in the hope that further information might be forthcoming concern- ing those whose names appear therein. That hope has received but slight fulfilment, and we once more call attention to the list in case there should be some who have not yet forwarded the information they may possess. This should be sent to Mr. Britten, 41 Boston Eoad, Brentford, Middlesex, who will also be glad to send a copy of the list to any who may not possess it. 57 NOTES ON BEITISH PLANTS. By C. E. Moss, D.Sc, F.L.S. I. SaGINA SAGINOiiDES. This species has come into some prominence recently owing to a discovery — or perhaps I should say a rediscovery — on Ben Lawers by the members of the International Phytogeographical Excm^sion, led by Mr. A. G. Tansley, in August, 1911. The particular Sagina which was then found and discussed has since been named S. scotica by one of the members of the party (Mr. G. C. Druce in Bot. Exch. Club. Brit. Eep. for 1911, iii. 14, 1912), and regarded as a hybrid of S. in-ocumhens and S. sagino'ides by two other members of the party (Dr. C. H. Ostenfeld in New. Phyt. xi. 117, 1912 ; Professor C. A. M. Lindman in Bot. Not. 267, 1913). So far as I understand the position of Ostenfeld and Lindman, these botanists regard the plant as a fixed and sterile hybrid, which originated long ago, and which has since continued to reproduce and spread itself by vegetative means. My own view of the plant is that it is a variety of S. sagino'ides. If Druce (in Journ. Bot. 1913, 91) is correct in identifying a specimen of Kobert Brown's as belonging to the disputed plant, then this is the earliest specimen known. I am not, however, very happy with regard to this determination. The specimen clearly belongs to the species S. sagino'ides, as I regard it ; but Brown describes his plant as " decandris " and " pentagynis " ; and one of the capsules of the specimen is nearly twice as long as the calyx. However, Druce is doubtless able to identify S. scotica. There is no date or locahty on Brown's label ; but another speci- men of *S'. sagino'ides on the same sheet was collected by Brown on Ben Lawers in 1794. The label of the plant mentioned by Druce contains a MS. description and a MS. name ; and the latter is, in my opinion, a MS. synonym of S. sagino'ides. I am assured that the disputed plant has for many years been definitely known to many Scottish botanists, and that these did not regard it as sufficiently different from S. sagino'ides to merit a special name. Lindman [op. cit.) gives some interesting details as to the history of the disputed plant in Scandinavia ; and the present note emphasises the fact that it was clearly known to Eeichenbach over seventy years ago, and known more or less clearly to several other Central European botanists in more recent years. It is not a rare plant in herbaria, where it is sometimes named S. ^;?-o- cumhens, but usually S. sagino'ides. Whilst the disputed Sagina was being discussed on Ben Lawers by the members of the International Phytogeographical Excursion, I dug up two sods of the plant. One of these I forwarded to Mr. E. W. Hunnybun, who drew the specimen for volume iii. of the Gamhvidgc British Flora; and the other I forwarded to Cambridge to be grown in my garden. Here it has flourished; and I have also been fortunate in successfully growing Journal op Botany. — Vol. 52. [March, 1914.] f 58 THE JOURNAL OP BOTANY specimens of the larger form of S. saginoules from the same locality. Hence I have had excellent opportunities of comparing the one with the other, and both with the alHed species S. pro- cumhens. In my opinion, the disputed Sagina has neither tlie appearance, nor the characters, nor the general behaviour of a hybrid. It is, in its essential features, constant over a very vast area in the northern hemisphere : in Scotland, it grows on several mountains in situations where its alleged parents are absent : the characters of the plant remain constant in cultivation : its pollen is normal : it produces, in abundance, plump capsules filled with good seeds ; and there is, so far as I can judge, no evidence of any factorial segregation. Under all these circumstances, I prefer to await the results of actual experiments before accepting the hypothesis, so ably maintained by Ostenfeld and Lindman, that the plant is of hybrid origin. As great emphasis has been placed upon the alleged infertility of the disputed plant, I repeat that in my garden it produced good seed freely throughout the summers of 1912 and 1913. In 1912, I sent samples of this seed to Druce and Ostenfeld. In early September, 1913, the Eev. E. S. Marshall and I visited Ben Lawers, and found that the little Sagina was quite fertile in its native haunts. Professor P. Graebner also states (fide Druce in Journ. Bot., loc. cit.) that the plant is fertile in the Botanic Garden at Berlin. I suggest that the apparent sterility of many herbarium specimens of this plant is due to their immaturity, and to their having been collected too early."'' This explanation is more especially likely to be correct when the "barren" herbarium specimens possess tetramerous flowers, for, as is shown later on, such flowers are common in the disputed plant in its early flowering stage. It is true that the disputed plant propagates itself very readily by vegetative means ; but all the British perennial members of the genus reproduce themselves more or less freely and effectively in what is essentially the same manner. The disputed plant is a little nearer S. procumhens than its ally, though it will be generally admitted that this fact does not demand the hypothesis of hybridity. A statement that the capsules of the disputed plant (" S. scotica Druce ") are larger than those of its near ally (" S. saginoules L.") is due to an accidental inversion of the names of the two plants {vide Journ. Bot. 142, 1913). The first published account of the two plants was given by Eeichenbach in his Icones Fl. Germ, et Helv. vol. v. (1841). Here both plants were named and figured. Eeichenbach placed them * A parallel case may here be mentioned. Salicornia fruticosa L. is very rarely found with ripe seeds on herbarium sheets : the seeds are not ripe until late October or November, when few collectors are at work. On the other hand, ripe seeds of S. glauca Del. are very common in herbaria : this species ripens its seeds nearly two months earlier than S. frnticosn, and at a time therefore when plant collectors are busy. SAGINA SAGINOIDES 59 both in his genus Spergella, a genus, I may add, which is rightly reduced to Sagina by all modern botanists. The disputed plant was there named Spergella saginoules, its larger ally Spergella macrocarpa. I do not detect the slightest confusion here. Eeichenbach, it seems to me, chose to regard the smaller of the two plants as the Linnaean type of the species ; and the larger plant he separated from it as a distinct species. The only differ- ence here between Eeichenbach and some other botanists is that Eeichenbach regarded the smaller plant as the Linnaean type, and others have so regarded the larger plant. There is nothing in the original Linnaean description to enable anyone to decide which of the two plants is really the Linnaean type : the specimen in the Linneean herbarium, whilst certainly belonging to the species S. sagmo'idcs as I understand it, is too young and too incomplete for me to state to which of the two forms it should be referred ; and therefore it seems to be essentially a case where the choice of the first author who separated the plants is binding. Some authors later than Eeichenbach have perhaps confused the issue by assuming that Spergella macrocarpa Eeichenb. w^as a larger plant than any described form of Sagina sagino'ides ; and perhaps this supposititious plant is the var. macrocarpa or the Sagina macrocarpa of some botanists. This, however, does not apply to all botanists who have taken up Eeichenbach's names ; and even if it did, it would not invahdate Eeichenbach's un- equivocal view of the case. Beck (Fl. Nied. Ost. 358, 1890) is perfectly clear about the matter, for his var. Jiiacrocarpa is definitely Spergella macrocarpa Eeichenb., and his var. typica is, by description, Spergella sagino'ides Eeichenb. In thus reducing Eeichenbach's two species to varietal rank, Beck has in my judgment correctly assessed the relationship of the two forms. Brilgger (in Jahresber. Naturf. Granbiind, xxiii.-xxiv. 71, 1881) has recorded a plant from the Bernina district of Switzerland as " Sagina saxatilis xprocumbens," adding in brackets after a short description and note " (S. media Brgg.)." The description given by Briigger is not a very satisfactory one ; and the only specimen so named in Briigger's herbarium at Chur is, as stated by Lindman {op. cit.), neither the disputed Sagina nor any other form of S. sagi- no'ides. Lindman's words (p. 273) are: — "Dr. Thellung has noted on the label [of Briigger's plant] that it is a common [form of] S. procumhens with some pentamerous flowers," and adds that he, Lindman, finds Thellung's identification to be "quite correct." However, from the evidence of unnamed specimens among Briig- ger's plants named S. procumhens, Lindman beheves (p. 274) that " there is in Briigger's herbarium quite sufiicient material of a ' Sagina media ' " ; and he accordingly adopts Briigger's name for the disputed Sagina. This name, I think, should be cited as " X S. media Briigger emend. Lindman." Wohlfarth (in Koch's Syn. ed. 3, i. 268, 1892) divides Sagina sagino'ides (sub nom. S. linncBV') into {a) var. micrantha and {h) * The point of view which I adopt with regard to the use of small letters for trivial names was stated in this Journal for 1913, p. 21. F 2 60 THE JOURNAL OF BOTANY var. decandra : these varieties date back to Ledebour's Fl. Eoss. i. 339 (1842). So far as I can judge, the var. decandra is S. sagino'ides as understood in the present communication ; and the var. mi- crantha (= Si)cr(jula micrantha Ledebour Fl. Alt. ii. 183, 1830) is a plant unknown to me. Lagerheim (in Kgl. Norske Vidensk Selsk Skr. for 1898, no. i, 4, 1898) found the disputed Sagina (i. e., S. sagino'ides var. typica Beck) in Scandinavia, and stated his view that the plant is a hybrid of S. 2^vociimbens and S. sagino'ides : Lagerheim's name may therefore be written x S. normanniana ; and this name is not open to the objection — an objection which applies to the name x S. media — that the author gave an inferior description and perhaps mixed or confused his own specimens of his own plant. It is worth noting that Swartz (in K. Vetensk. Acad. Handl. 44, 1. 1, fig. 2, 1789) describes his Spergula sagino'ides as possessing only five stamens. Smith draws attention to this statement — exemplified also in Swartz's figure — when describing his own plant and figure (Eng. Bot. ed. 1, t. 2105), and adds that whether his (Smith's) plant is Swartz's or not it is certainly Linne's. As Presl cites Smith's figure when founding his Sagina linncBi, we have, in this remark of Smith's, the origin of Presl's trivial name. In all the specimens of Sagina sagino'ides which I have observed, including both the forms here discussed, the stamens have been n + n, where n is four or five ; and Presl correctly gives the number of the stamens of the species as eight or ten. I do not doubt that Presl deliberately included both forms of the species in his S. linncei, for tetramerous flowers in the larger form are quite rare. It is also fair to say that Smith also included both forms in his Spergula sagino'ides, for both forms are so named in his herbarium. From the point of view, then, adopted in the present paper, the following citations are set out : — Sagina saginoides Dalla Torre Anleit. Beob. Alpenpfl. 75, in Hartinger's Atlas der Alpenfl. (1882) incl. Sagina macrocarpa ; Britton in Mem. Torr. Club. v. 151 (1894) ; Spergida sagino'ides L. Sp. PI. 441 (1753) ! ; Smith Fl. Brit. 504 (1800) ! ; Sagina linnai C. B. Presl Eel. Haenk. ii. 14 (1831) ; Sagina saxatilis Wimmer Fl. Schles. 75 (1841). (a) S. SAGiNOiDES var. macrocarpa Beck Fl. Nied.-Ost. 358 (1890) ; Spergella macrocarpa Eeichenbach Icon, v., 26, fig. 49636, (1841) ; Sagina macrocarpa Maly Enum. PI. Austr. 293 (1848) ; Sagina saxatilis var. macrocarpa Hausmann Fl. Tirol. 133 (1854). Icones : — Eng. Bot. ed. 1, t. 2105, as Spergula sagino'ides ; Fl. Dan. t. 1577, as Spergula sagino'ides ; Svensk Bot. t. 765, as Spergida sagino'ides ; Eeichenbach Icon. v. t. 202, fig. 49636, as Spergella macrocarpa. Exsiccata : — Billot, 1423 (partim), as Sagina linnai; Fellman, 42, as Arenaria hiflora (corrected to ;S'. saxatilis) ; Fries, ix. 40, as 5. saxatilis. (b) S. SAGINOIDES var. TYPiCA Beck loc. cit. ; Spergella sagi- no'ides Eeichenbach loc. cit. t. 4962 ; x Sagina normanniana SAGINA SAGINOIDES 61 Lagerheim loc. cit. ; S. glabra var. scotica Druce in New Phyt. X. 325 (1911) ! ; S. scotica Druce in Bot. Exch. Club Brit. Rep. for 1911, iii. 14 (1912) ! ; S. procwnbens x sagino'ides Ostenfeld in New Phyt. xi. 117 (1912) !, ? excl. syn. Briigger ; Lindman in Bot. Notiser. 267, et fig. (1913) !, ? excl, syn. Briigger ; x S. media Briigger [loc. cit.] emend. Lindman oj;. cit. p. 273. Icones : — Reichenbach oj). cit. t. 4962, as Spergella sagino'ides. Exsiccata : — Billot, 1423 (partim), as Sagina linncei ; Reichen- bach, 1095, as Spergella sagino'ides ; Schultz et Winter, i. 21, as Sagina linnai. In citing Billot's No. 1423 (as exemplified in Herb. Univ. Cantab.) under both varieties, I am casting no reflection on his perspicacity. I think it is quite probable that he deliberately in- tended in these specimens to indicate his view of the species Sagina sagino'ides ; and, if this is really so, it only remains for me to add that I follow Billot (as well as Smith and Beck and probably Presl) in my concept of the species in question. The distinguishing characters of the two British varieties of Sagina sagino'ides are tabulated below. These characters are taken from fresh specimens grown in my garden originally from Ben Lawers, from fresh material growing on the same mountain, and from Lindman's excellent description [op. cit.). It is well to add that, as yet, cultivation of my own garden specimens has induced no alteration worth mentioning in the characters of either variety, and that Lindman (p. 272) seems to have detected no differences in Scandinavian material which had been in cultivation for fifteen years. S. sagino'ides var. typica. Habit Less robust and more strag- gling. Vegetative propagation by axillary buds or shoots more pronounced. Barren rosettes More numerous, but indi- vidually smaller. Leaves short- er (up to about 1-8-2-0 cm. long). Pedicels More slender. Floivers First flowers mostly tetra- merous. Later flowers mostly pentamerous. Latest flowers mostly tetramerous.''' S. sagino'ides var. jnacrocarpa. More robust and less strag- gling. Vegetative propagation less pronounced. Less numerous, but indi- vidually larger. Leaves longer (up to about 2-5 cm. long). Less slender. Usually pentamerous, very rarely tetramerous.''' * In this connection, it should be mentioned that S. procumhens is some- times pentamerous even when growing in lowland localities where any form of S. sagino'ides is unknown. 62 THE JOURNAL OF BOTANY S. sagino'ides var. macrocarim. On the average, about 1*6 mm. long and 2-25 broad. Usually erect (or suberect) in fruit, rarely spreading. On the average, about 1'5 mm. long and 2*0 broad.'-' On the average, about 5-8 mm. long, and about 1'3-1"9 times as long as the calyx. S. sagino'ides var. iyinca. Sepals On the average, about 1*3 mm. long and 2'3 broad. Occasionally spreading in fruit, often erect (or suberect). Petals On the average, about 1-0 mm. long and 1-5 broad.''' Capsides On the average, about 3-5 mm. long, and about 1-1-1-3 times as long as the calyx. I feel that it cannot be successfully maintained that the characters tabulated above are of sufficient importance to justify botanists in regarding the two plants as specifically distinct. It will be noticed that the differences are either comparative or refer to matters of mere number and size. I find that, when large numbers of specimens of the two plants are compared, the actual discontinuity between the two varieties is trifling, though I admit that with patient examination it is possible to determine precisely almost any complete and mature specimen whether living or dried. Further, if Druce is correct in identifying Kobert Brown's plant, alluded to earher on in this paper, as /S. sagino'ides var. typica (i.e., S. scotica Druce), then there is not merely an absence of discontinuity in the important character of tlie size of the capsule, but there is actually a considerable amount of overlapping. S. sagino'ides, in each of its British forms, may be distinguished from the allied S. procumbens by its more robust habit, its larger leaves which are scarcely mucronate, its usually longer pedicels, its more frequently pentamerous flowers, its larger petals, its larger capsules, its usually erect (or suberect) fruiting sepals, and its much greater abundance in sub-alpine and truly alpine localities, usually in wet situations. From the allied S. subulata it may be distinguished by its being totally glabrous and eglan- dular, by the much less pronounced apical mucronation of the leaves, and by its smaller flowers and capsules. It has to be confessed, however, that " species " in this genus have been made exceedingly small, and that reduction in other parts of the genus is desirable ; for example, are not S. ciliata and even S. reuteri too closely allied to S. apetala to be allowed to remain as separate species ? The two varieties of S. sagino'ides can scarcely be said to grow in distinctive habitats. In Scotland, both occur in and near sub- alpine and alpine streamlets and springs. The var. typica tends to spread from the springs and streamsides on to the surrounding siliceous grassland, and even on to small hillside Iclgers, more than * Petals are frequently present in S. procumbens, but are not more than half as long as the sepals. SAGINA SAGINOIDES 63 does the var. macrocarpa ; but the two often occur side by side in the wetter sitviations. The var. typica descends to lower altitudes (ca. 550 m.), and the var. macrocarpa to higher altitudes (ca. 1320 m.); but the stations overlap a great deal.* The latitudinal range of the two varieties is practically identical, as the var. typica occurs in Iceland, Scandinavia, Central Europe, Asia, North America, and Greenland, as well as in Scotland (see Lindman op. cit). As to the respective names the plants should bear if they are regarded (erroneously regarded, as I think) as different species, I submit that Eeichenbach's names have not, in recent discussions, received due recognition. The trivial names of Keichenbach are legally correct : he was the first author to name the two plants as separate species : his figures are less faulty than those of his predecessors : there is nothing in the original descx'iption to invalidate his choice of the type : there is no doubt or ambiguity or confusion regarding the identity of his plants : later authors, including such an eminent systematist as Beck von Mannagetta {loc. ci^.), have accepted his names ; and hence I can see no valid objection whatever (supposing the two plants are regarded as different species) against the names Sacjina macro- carpa (Maly, op. cit.) for the larger plant and Sagina sagino'ides (Linn, sub nomine Spcrgula, emend. Eeichenbach sub nom. Sper- gella) for the smaller plant. If it be contended that the name Sagina sagino'ides has under the new circumstances become a nomen confusum, then I think the name S. scotica (Druce, loc. cit.) holds the field ; but I should not agree to the rejection of a Linnaean trivial name where the confusion is so slight as in this case. If, indeed, such a contention were upheld, how many LinnaBan names, either generic or specific, would remain valid ? Whilst discussing this nomenclatorial matter, one may perhaps be pardoned for alluding to the form of such names as Sagina sagino'ides. By the International Eules, one has to reject such names as Castanea Castanea ; but it must be admitted that names of the latter form, whilst doubtless objectionable from some points of view, are less nonsensical than such names as Sagina sagino'ides and Gerastium cerastio'ides. The actual outcome of the whole discussion is that two very closely allied plants, properly distinguished and figured by Eeichenbach more than seventy years ago, have been brought into prominence, and their characters and distribution worked out in some detail. It is also clear that, even with precisely the same critical plants before them, systematic botanists cannot always agree as to the characters, the nature, and the status which these plants actually possess ; and until agreement has been reached regarding the status or rank of a plant, it is, in the nature of the case, quite impossible to attain unanimity with regard to the name it shall bear. * S. procumbens is recorded by White (Fl. Perthshire, p. 86) up to 1000 m. ; but all these higher altitudes attributed to S.i^rocnmbens now need revision, as this species has been confused with S. sagino'ides var. typica. 64 THE JOURNAL OF BOTANY THE MUSS^NDAS OF MADAGASCAE. By H. F. Wernham, B.Sc. (Department of Botany, British Museum.) In a recent paper (Journ. Bot. 1913, p. 233) I pointed out the possible significance of the relative size of the limb and tube of the corolla in the genus Mussanda. In the case of species with one of the calyx-lobes developed as a foliar attractive organ, the limb is relatively insignificant. In the case of the Mussanda species found in Madagascar and the Mascarenes — all endemic save one [M. arcuata) — the corolla-limb is relatively large, and the calyx-lobes subequal {^Landia), none being amplified into the foliar organ characteristic of, e. g., M. luteola. M. elegans is the only continental African species known in which the corolla-limb is amply developed and the calyx-lobes invariably subequal ; and it is thus worthy of note that the two conditions of the corolla-limb in question are in significant corre- spondence with two respective areas of distribution, viz., con- tinental Africa and the islands in question. A further interesting point in the same connection is the fact that M. arcuata, which is transitional from the aspect of the corolla-limb and the calyx- lobe development (Journ. Bot. 1913, p. 233), occurs widely in both the areas named. Again, all the species now concerned appear to be erect in habit, either shrubs or trees, with the exception of M. arciiata. All the eighteen species dealt with in the present paper are endemic in Madagascar, with four exceptions, namely, ill. arcuata, Poir., just mentioned, which occurs in Mauritius and Bourbon, as well as in Madagascar ; M. mauritiensis (p. 66) and M. Stadmannii Michx. (p. 66), confined to Mauritius; and 71/. Landia Poir., ex hb. Justice Blackburn, the locality of which is in doubt, but pro- bably Mauritius. Key to the Species. Plants glabrous, or almost so, including the flowers ; corolla- tube, 2"5 cm. long at most, usually much less, occasionally puberulous externally along 5 obscure ridges 1. arcuata. Plants not glabrous. Corolla-tube over 3 cm. at least, usually over 4 cm., never glabrous. Pubescence on principal leaf-veins adpressed, or mostly so. Calyx-lobes + rigid, appreciably broad, not subsetaceous. Leaves not glabrous above. Leaves not oblanceolate (except in M. asperula) ; + rough above with very short hairs, usually much over 10 cm. long. Leaves broadly oval, shortly acuminate at most, barely twice as long as broad. Corolla-tube about four times as long as lobes 2. tricJiophlebia. Corolla-tube less than three times as long as lobes. Calyx -lobes linear, mostly upwards of 1 cm. long 4. Stadmannii. THE MUSS^NDAS OP MADAGASCAR 65 Calyx-lobes ovate-lanceolate, barely 5 mm. long usually o. Landia. Leaves long-acuminate, mostly over three times as long as broad 5. mauritiensis. Leaves rather small, mostly broadly oblanceolate, about two and a half times as long as broad 6. asperula. Leaves oblanceolate, sparsely pilose above with longish red hairs, barely 5 cm. at longest 11. ramosissima. Leaves glabrous above, or punctate, except for obscure, closely adpressed pubescence on veins. Corolla-tube over 6 cm. long 8. erectiloba. Corolla-tube not much over 4 cm. long. Leaves punctate above ; calyx-lobes mostly less than 1 cm. long 9. punctata. Leaves not punctate above ; calyx-lobes nearly 2 cm. long in flower 10. hymenopogonoitles. Calyx-lobes subsetaceous for greater part of their length. 7. Pervillei. Pubescence on principal veins below spreading. Flowers not sessile. Fruiting calyx-lobes attaining 2-2-5 cm. in length. Leaves less than 8 cm. long 12. arachnocarpa. Leaves over 15 cm. long 16. lyilosa. Fruiting calyx-lobes less than 1 cm., rarely over 1 cm. long, then setaceous. Leaves markedly scabrid above, with short hairs swollen at base 17. scabridior. Leaves not scabrid above. Leaves less than 6 cm. x 2-5 cm 13. Humhlotii. Leaves over 12 cm. x 6 cm 18. vestita. Flowers sessile or subsessile, solitary, or in a few-flowered head. Corolla-tube less than 5 cm. long, lobes less than 2 cm. 14. inonantlia. Corolla-tube over 8 cm. long, lobes nearly 3 cm. 15. fusco-pilosa. Note. — The species are arranged as far as possible in order of their affinity. 1. M. ARCUATA Poir. in Lam. Encycl. Meth. iv. 392 (1795). Landia stelligcra et L. astrograiilia Comm. ex Hb. Mus. Paris. Nom. vulg. " Caca poule" (Bouton). Madagascar : Forbes ! Gerrard, 18 ! Baron, 6822 ! Ilumblot, 197! Lijall, 288! Central: Baron, 570! 2472! 3692! 4767! Nossi-be : seashore, Hildehrandt, 2911 ! Mbatourana, between Tamatave and Antananarivo : Meller ! S. Marie Is. : Forbes ! Mauritius : Ayres ! Bojer ! Bouton ! Garmichael, 9 ! Graham ! Hilsenbenj £ Bojer \ Roxburgh I Sieber (Fl Maur.), 78! Thomp- son ! Bourbon : Aublet ! Balfonr ! Commerson 1 Hardioicke ! The above do not differ essentially from the Continental speci- mens (1913, 274). One or two of the Mauritius specimens show signs of a pilose midrib on the under side of the leaves and of hairs in lines on the corolla-tube ; but none can properly be classed with my variety /i jmbescens (I. c. 274). 31. arcuata is strikingly distinct from all the other Madagascar species in its glabrous shoots, leaves and small flowers, as well as 66 THE JOURNAL OF BOTANY its scandent habit. Bouton remarks of this " arbrisseau liane " that it grows in the ravines in every part of the island of Mauritius, and flowers almost all the year round. He refers to it as a " plante medicinale." 2. M. TRiCHOPHLEBiA Baker in Journ. Linn. Soc. xx. 66 (1883). M, ')nacro2)oda Baker, loc.cit. xxi. 410. Central Madagascar : Baron, 497 ! 1764 ! 2181 ! 2293 ! 3088 ! 3642 ! 3644 ! 3974 ! 7017 ! Betsileo-land ! Baron, 107 ; East Tmerina, Andrangoloaka, Hildebrandt, 3615 ! Hbb. Mus. Brit. Kew. 3. M. Landia Lam. 111. t. 157, 2 ; Poir. in Lam. Encyc. iv. 392. M. holosericea Sm. in Eees, Cycl. xxiv. n. 6 (1819). M. latifolia Poir. Encycl. Suppl. iv. 36. Mauritius ? Hb. Blackburn. Hb. Kew. In the absence of Poiret's type and the original of Lamarck's figure {supra), I find considerable difficulty in identifying this species. According to the figure in question, the leaf-bases are noticeably rounded, and the ovate calyx-lobes and ovary short. The only specimen I can find in the British herbaria which matches this figure at all satisfactorily is the one quoted, origi- nating from the herbarium of Mr. Justice Blackburn, now pre- served at Kew. This, collected probably in Mauritius, I am disposed to regard as the true M. Landia Poir. This is distinguished readily from the two succeeding species in size of calyx and ovary on the one hand {M. Stadmannii), and in leaf-shape on the other {M. mauritiensis). The confusion which has arisen among these three species is not improved by the fact that De Candolle (DC. iv. 372) gives M. Landia Sm. in Eees, Cycl. xxiv. as a synonym of M. Stadmannii DC. Smith's M. Landia is certainly the same as M. Landia Poir. and M. Landia Lam. 1. 157, for Smith quotes this figure under his M. Layidia, and Poiret's description refers to the same plant. 4. M. Stadmannii Michx. ex DC. Prod. iv. 372. Oxyanthus cymosus Keichenb. in Sieb. M. Maur. {v. infra) ex DO. Prod. loc. cit. Mauritius : Sieber ii. 79 ! Bojer ! Bouton ! Carmichael, 165 ! Bouton's label reads : " Vulg. Quinquina rudigera. Croit dans les forets dans tous les quartiers de File. Arbrisseau se couvrant de fleurs blanches. Fl. mois de Fevrier." This has been confused with M. Landia (q. v.) ; but it seems to be quite distinct in its long, sublinear calyx-lobes and its elongated ovary. 5. M. mauritiensis Wernham, sp. nov. Eamulis sericeis ; foliis lanceolatis, longe acuminatis, basi acutis, apice acutissimis, supra asperulo-pubescentibus, subtus in venis sericeo-pubescen- tihus, 2)etiolo dense flavo-sericeo; floribus . . . fructibus oblongis sparse, pedicellis densius sericeis, corymbose dispositis ; calycis limbo lobis triangulari-lanceolatis acuminatis acutis dense sericeis coronatis. Mauritius: "in sylvis, ad radices montium," Bojer \ " Sur les hautes montagnes," Hb. Blackburn ! Hb. Kew. THE MUSS^NDAS OF MADAGASCAR 67 Eeadily distinguished, even in the absence of flowers, from the previous species and M. Landia by the lengthily acuminate, large lanceolate leaves. The latter measure 14-22 cm. x 2-3-6 cm. ; secondary veins 8-10 pairs ; i^etiole to 1'8 cm. Peduncle 2*5-3 cm. Calyx-lohe^ 5-7 mm. Fruit 1*7 cm. long, 8-5 mm. wide. 6. M. asperula Wernham, sp. nov. Eamulis novellis densi- uscule appresse pubescentibus ; foliis oblanceolatis v. anguste obovatis, basi acutis apice breviter acuminatis acutis, supra scabridulis, subtus in venis primariis strigillosis in reticulo minute asperulo-pubescentibus aliter glabris, petiolo brevi, stijyuUs parvis lanceolato-triangularibus alte bifidis acutis insuper subsetaceis ; inflorescentia bis trichotome corymbosa ramulis appresse pubes- centibus, hracteis bracteolisque setaceis ; calycis lobis lanceolatis acuminatis acutis appresse pilosis ; corolla tubo extus breviter sericeo insuper parum ampliato, lobis ellipticis mucronatis ; ovario oblongo sericeo. Central Madagascar : Baron, 493 ! Hb. Mus. Brit. Near 71/. erectiloba Wernham, differing in the shape and indu- mentum of the leaves, &c. Leaves 5-5-9'5 cm. x 2-2-3"3 cm. ; petiole less than 1 cm. ; secondary nerves, 7-9 pairs ; stipules 7 mm. long at most. Peduncles, primary, 3 cm. or longer ; secondary, 2*5 cm. ; bracts 8 mm. long. Pedicels, 1-6 mm. Calyx- lobes to 7 mm. Corolla-tnhe 5-5 cm., lobes 1*4 cm. x 4-6 mm. Ovary 6 mm. long. 7. M. Pervillei Wernham, sp. nov. Eamulis obtuse quad- rangularibus, novellis la3viter dilute brunneo-sericeis ; foliis amplis ovalibus breviter acuminatis acutis, supra asperulo-pubescentibus subtus in venis conspicuis sericeo-strigosis, reticulo tertiario fusco manifesto, petiolis mediocribus nonnunquam longiuscuHs sa^pius appresse sericeo-brunneo-pilosis, stipulis rigidiusculis a basi lato triangularibus setaceo-acuminatis extus sericeis ; florihus in pani- culis corymbosis multifloris subtrichotomis ramulis appresse pilosis, hracteis setaceo-linearibus ; calycis lobis subsetaceis ; corollcB extus primo densiuscule sparsiuscule tandem appresse pubescentis tubo gracili insuper parum ampliato, lobis ellipticis mucronatis ; ovario breviter strigilloso tardius glabriore, in bacca tandem breviter oblonga subglabra a calycis lobis setaceis coro- nata pedicellata maturante. Madagascar : North and north-west : Baron, 6373 ! 5800 ! Perville ! Nossi-b6 : Hildebrandt, 3003 ! Leaves 10-17 cm. x 5-5-9'5 cm. ; petiole l-2'5 cm. ; secondary veins 10-14 pairs ; stipules 1 cm. long, 5 mm. broad at base. Flowering peduncle about 2-5 cm. ; primary bracts about 8 mm. GalyxAohe's, to about 8 mm., somewhat accrescent in fruit. Corolla-tuhe 3-5 cm., or rather longer ; lobes about 1 cm. x 5 mm. Fruit 1-7 cm. long, 8-9 mm. wide. 8. M. erectiloba Wernham, sp. nov. Arbor, ramulis sparsim et obscure appresse pilosis mox glabrescentibus ; foliis ellipticis v. oblongis plerumque brevissime acuminatis basique apiceque acutis, supra Isevibus nisi in venis hie inde pilis brevibus appressis indutis 68 THE JOURNAL OF BOTANY glabris, subtus in venis sparsim appresse pilosis, petiolo saepius breviusculo similiter induto, stipidis basi brevi lato oblongo insuper bifidis subulate- v. subsetaceo-linearibus sparsim appresse pilosis ; injiorescantia trichotome corymbosa ca. 9-flora, bracteis setaceis, ramulis subglabris complanatis ; calycis lobis rigidiusculis a basi anguste triangulari linearibus subglabris ; corollcB tubo longo gracili insuper parum ampliato extus breviter sericeo, lobis lanceo- latis acuminatis acutis ; ovario anguste infundibulari basi attenuate sparsim strigilloso. Madagascar : Ankafana, Deans Coivan ! Tanala, Ambobimi- tombo forest, 4390-4680 ft., Forsyth Major, 274! Hbb. Mus. Brit., Kew, Allied to M. trichophlebia Baker, but differs especially in the shape and indumentum of the leaves and calyx-lobes ; and more nearly, perhaps, to M. Pervillei Wernham, from which it differs in the size of the corolla and the shape of the lobes. Leaves 6-5-8*5 cm. x2-3-3-5 cm., petiole to 1 cm., secondary veins 8 pairs ; stipule entire, basal part 3-4 mm., with two distant prongs 8 mm. long. Primary peduncle 2 cm., secondary yeduncles 1'3 cm. long, each commonly bearing 3 flowers with pedicels about 8 mm. long. Caly x-lohe^ 1-1-5 cm. long. Corolla- tube 6'5-7 cm. long, lobes 1-6 cm. x 4-5 mm. Ovary 7 mm. long. Var. 13 scABHELLA Wernham, var. nov. Foliis amplioribus supra scabridulis subtus in venis strigillosis ; corolla minus valida. Madagascar : Fort Dauphin, Scott Elliot, 2607 ! Chisel, 97 ! Hbb. Mus. Brit., Kew. Cloisel writes on his label : " Tatome, grand arbre a fleur rouge et blanc." 9. M. PUNCTATA Drake, Hist. PL Madag. t. 447 (1897). Frutex nisi floribus subglaber ; foliis plerumque oblanceolatis brevissime acuminatis acutis, supra trichomis minutis basi inflatis hie inde conspersis punctatis, subtus in venis sparsiuscule brevissime appresse strigillosis, alitor glabratis, petiolatis, stipidis triangu- laribus bipartitis acutis ; floribus per 1-3 in axillis imis et caulis furcis dispositis ; calycis lobis linearibus nonnunquam inaequalibus extus qua ovarj2i7;i subcampanulatum ferrugineo-sericeis; corollce breviusculiE tubo extus rufo-sericeo insuper sensim ampliato, lobis oblongo-ovatis submucronatis. Madagascar: Mahalougouloue ('?), Thovipsonl Hb. Mus. Brit. Remarkable for the almost glabrous, punctate leaves, and the rather short flowers considerably widened above. As in the case of other Rubiaceae figured so excellently by d'Apreval for Drake's unfinished work, no description, unfortunately, is extant; but I have no doubt whatever that Thompson's plant is specifically identical with M. punctata as figured. Leaves not exceeding 8-5-9 cm. X 3-3-3 cm., petiole 1-1-5 cm., 8-11 pairs of secondary veins. Peduncle 0-10 mm. Pedicel 1 cm. or more. 6'a/^a;-lobes usually 1 cm. or slightly longer. Gorolla-twhe 4 cm. long, widen- ing, from a distance of about 2-5 cm. above the base to nearly 1 cm. broad at the mouth ; lobes 8-5 mm. x 5 mm. 10. M. HYMENOPOGONOiDES Baker in Journ. Bot. xx. 138 (1882). THE MUSSiENDAS OF MADAGASCAR 69 Central Madagascar: Tanala, in forest, Baron, 313! Eemarkable for the narrow leaves, quite glabrous except for the relatively scanty and closely adpressed hairs on the veins below^, and for the long narrow calyx-lobes 1-6 cm. or longer in the flower, and increasing to over 2 cm. in the fruit. The whole plant approaches glabrousness, such indumentum as there is being very closely adpressed ; the parts thus present a remarkably smooth appearance. 11. M. ramosissima Wernham, sp. nov. Verisimiliter frutex erectus, ramosissimus, ramulis novellis densissime appresse rufo- pilosis cortice tandem cinereo rugosulo indutis ; foliis parvis oblanceolatis apice subrotundatis nee acuminatis, utrinque praesertim in venis ac subtus densius appresse rufo-pilosis, petiolo dense rufo-sericeo ; stipulis lanceolatis bipartitis acu- minatis acutis ; florihus in foliorum ramulorum apicem versus confertorura axillis dispositis, longiuscule pedicellatis ; calycis lobis oblongo-linearibus subacutis dorso in costis 3-5 prominenti- bus rufo-strigosis ; corollcB extus densiuscule sericeoe tubo gracili parum insuper ampliato lobis elliptico-ovatis breviter cuspidatis ; ovario angusto in pedicellum insensim desinente rufo-strigoso a calyce manifesto superato. Madagascar : Hnmblot, 392 ! Hb. Kew. A very distinct species, readily distinguished by the shape of the small, red-haired leaves and the long slender flowers, &c. The affinity is with M. imnctata Drake, and M. hymenopogoiioides Baker. Leaves 4*5 cm. long, 1-7 cm. broad in upper part, with stalk as much as 6 mm. long and 6 secondary veins, prominent below, on either side of the midrib; stipules 5-7 mm. x 2-5-3 mm. Pedicel 1 cm., passing into ovary about the same length. Calyx- lobes 1-2 cm. x 1"7 mm. Gorolla-tuhe 5-5 cm. long, lobes 1"3 cm. X 6-7 mm. 12. M. arachnocarpa Wernham, sp. nov. Frutex v. arbor, ramulis sparsiuscule hispidulo-rufo-pubescentibus ; foliis chartaceis oblanceolatis basin cuneatum versus leniter angustatis breviter acuminatis acutissimis, supra in vena centrali impressa minute patento-strigillosis alitor hispidulo-pubescentibus, subtus praesertim in venis prominulis moUiter asperulo-pubescentibus, petiolo brevis- simo pubescente ; stipulis binis subulato-setaceis rufo-pilosis ; florihus in umbellis paucifloris terminalibus dispositis ; bracteis bracteolisque setaceo-linearibus ; fructu ellipsoideocostato sparsius- cule rufo-pubescente ; pcdicellis longiusculis ferrugineo-pubescenti- bus ; calycis lobis rigidiusculis linearibus subacutis velut bacca indutis. Madagascar: Fort Dauphin, Scott Elliot, 2624! Hb. Kew. Fr. May. The specimen, unfortunately, bears no flowers ; but it clearly represents a distinct species allied to M. Huviblotii, from which it differs in the shape and indumentum of the leaves, the stipules, &c. Leaves 5*5-7 cm. x 1*5-2 cm. ; petiole barely 5 mm. at most, secondary veins 8-11 pairs ; stipules 9 mm. Bracteole (fruit) 6 mm. Pedicel (fruit) 1-2 cm. Fmit 1'5 cm. long, rather more 70 THE JOURNAL OF BOTANY than 1 cm. wide, crowned by the persistent calyx-lohes, attaining nearly 2 cm. in length. 13. M. Humblotii Wernham, sp. nov. Verisimiliter frutex erectus, i-amulis novellis densiuscule ferrugineo-pubescentibus pilis patentibus, tandem cortice sublsevi nee rugoso indutis ; foliis crassiusculis elliptico-oblongis vix aeuminatis acutis basi obtusis v. subrotundatis, supra in vena centrali impressa rufo-strigosis aliter breviter sparsim strigillosis, subtus in venis prominentibus prseser- tim in centrali densiuscule aliter breviter et obscure patule pubes- centibus, petiolo brevi dense ut costa induto ; stijmlis parvis triangularibus acutis setoso-acuminatis bipartitis ; florihus pedi- cellis longiusculis dense rufo-pubescentibus in summis axillis solitariis ; calycis lobis linearibus subacutis rigidiusculis sparse pilosis ; corolla tubo extus pi'aecipue insuper densiuscule basin versus sparsiuscule plus minus patente pilosis, insuper nee multo ampliato, lobis ovatis breviter aeuminatis acutis extus rufo-sericeis intus minutiuscule pubescentibus ; ovario dense ferrugine sub- tomentoso. Madagascar : Humblot, 617 ! Hb. Kew. Allied to M. ramosissima, but distinct in the soft, spreading pubescence, the leaf-venation, and the size of the flowers. Leaves 3'5-4-5 cm. x l'5-2 cm. ; petiole 6 mm. at most ; secondary veins 6-9 pairs ; stijndes not more than 6*5 mm. Pedicels barely 1 cm. at most. Calyx-lohes, to 1*5 cm. long. Co?-o//a-tube about 3"5 cm. long, lobes 1*4 cm. x 6"5 mm. Ovary 6 mm. long. 14. M. monantha Wernham, sp. nov. Frutex ramulis novellis densissime rufo-pilosis; foliis ellipticis v. oblongis utrinque angus- tatis et acutis, supra in vena centrali densiuscule aliter sparsim tamen uniformiter rufo-strigosis margine ciliatis, subtus discolori- bus in venis dense aliter sparsim patente pilosis, petiolo brevi piloso ; stijndis membranaceis hirsutis triangularibus insuper in setis 2 divisis ; florihiis suaveolentibus candidis solitariis sessilibus V. subsessilibus 5-6 meris ; calycis lobis linearibus nonnunquam setaceo-subulatis inter longiores acutis ; corolla tubo insuper nee multo ampliato extus patente hirsute, lobis ellipticis breviter aeuminatis acutissimis extus sparsiuscule appresse pilosis ; ovario dense ferrugineo-piloso. Madagascar : TJiompson ! in Hb. Mus. Brit. Between Tamatave and Antananarivo, on clay at 3000 ft., Mellerl in Hb. Kew. An isolated species, with M. fusco-inlosa Baker, perhaps, as its nearest ally, but with many obvious differences from this ; it is especially remarkable for the solitary sessile flowers. Leaves 7-11*5 cm. x 3-5-4-8 cm., petiole barely 1 cm. at most, stipules 8 mm.; secondary veins 8-11 pairs. Calyx-lohes 1'8 cm. long. Corolla-inhe upwards of 4-5 cm. long; lobes 1'3 cm. x 7 mm. Ovary 5-6 mm. deep. 15. M. Fusco-PILOSA Baker in Journ. Linn. Soc. xxi. 410 (1885). Central Madagascar : Sarow, 2467 ! 2470! 6118! THE MUSS/ENDAS OF MADAGASCAR 71 One of the largest-flowered species known, with corolla-tube nearly 9 cm. long, readily distinguished by the size and long, spreading indumentum of its sessile flowers. 16. M. PiLOSA Baker in Kew. Bull. 1895, 105. North Madagascar : Baron, 6179 ! Hb. Kew. Distinguished by the dense spreading hairs on the narrow, oblanceolate leaves and the branchlets, and by the long linear calyx-lobes, nearly 2-5 cm. long in the flower. 17. M. scabridior Wernham, sp. nov. Eamulis hispidulo- molliter pubescentibus ; foliis inter majora, supra pilis brevibus curvatis basi inflatis scabris subtus breviter praesertim in venis hispidulis ; stiiyulis fere ad basin in setis 2 breviusculis distantibus divisis ; in/lores cent ia multiflora corymbosa, ramulis hispidulo- pubescentibus, bractcis parvis setaceis ; calycis lobis lanceolato- linearibus, in fructu parum accrescentibus ; bacca oblonga costata a calycis limbo persistente coronata. Central Madagascar : Baron, 1505 ! 3975 ! Hb. Kew. With affinities to M. vestita this is readily distinguished from all the other species by the markedly scabrid upper surface of the leaves. These are elliptical, reaching about 14 cm. x 8'5 cm., with pubescent stalk as much as 2 cm. long ; there are 8-11 secondary veins on either side of the midrib; they are but shortly acuminate, with obtuse apex. 18. M. VESTITA Baker in Journ. Linn. Soc. xvi. 166. Madagascar : Hilsenberg it Bojer ! Central : Baron, 3731 ! Betsileo-land, Baron, 55! Langley-Kitchmgl Hbb. Mus. Brit., Kew. Distinguished by the thick, velvety, light-coloured tomentum on the under surface of the leaves. According to Hilsenberg and Bojer the native name is Fatoora. P MACEOCALYX Wernham, var. nov. Calycis necnon corolla) lobis longioribus, hujus tubo loborum longitudinem bis paullo excedente. Central Madagascar : Anfakana, Deans Cowan I Mt. Antety, above Ambositra, Forsyth Major, 635 ! Hbb. Mus. Brit., Kew. Calyx-lohes 1 cm. or longer ; coro/Za-tube 3-5 cm. long, lobes 1-5 cm.-l'8 cm. x 9 mm. Species dubi^. M. EEiANTHA Rich, in M6m. Soc. Hist. Nat. Par. v. 246 (1829). The description is very inadequate. The indumentum of petioles, branchlets and subulate stipules is described as " rufo-sericeis " ; and the calyx-lobes are lanceolate and equal ; so that the species belongs to the section Landia. The subulate stipules should be distinctive, as also the silky, presumably adpressed, indumentum. M. Thouarsiana Baill. Adans. xii. 295 (1879). Based on a plant gathered by Dupetit-Thouars in Madagascar. Described as glabrate, with leaves sometimes irregularly denticulate on the margin ; inflorescence very dense and much branched. These characters are unfavourable to the inclusion of this plant in the 72 THE JOURNAL OF BOTANY genus MusscBmla, as is also the presence of a spurious wing to the seeds. The corolla is unknown ; and the author himself expresses a doubt as to the genus. M. ? ciTRiFOLiA Lam. Encyc. iv. 393 (1797). Nom. vulg. Charro, Eees Cycl. xxiv. According to the description, the leaves are borne 3 at each node, the calyx and corolla are glabrous, and the latter is small. These characters suggest that this plant, which was collected by Martin in Madagascar, should not be included in this genus ; and a similar conclusion applies to the so-called nearly allied M.? longifolia Lam., I.e. M, DISCOLOR Thouars, in Eoem. & Schult. Syst. v. 254 (1819)- Another imperfect description. Baker (Hb. Kew) is of opinion that this may be identical with his M. vestita (q.v.) ; and in view of the description, " foliis . . . subtus piloso-canescentibus," this is not unlikely. As a matter of fact, the very locality is doubtful. THE MOSS FLOKA OF SUFFOLK. By Arthur Mayfield. The following list of Suffolk mosses is intended to supplement the two papers published by the Kev. E. N. Bloomfield in this Journal for 1885 (pp. 233-238) and 1888 (pp. 69-71). During 1912-13 I have been gathering mosses in the county, chiefly in the parish of Mendlesham, near the centre of the county, where the subsoil is mainly the chalky boulder clay. I have thus been able to add considerably to the East Suffolk list, and also to record the occurrence of plants which either had escaped notice or were not considered species by the earlier botanists. Mr. Bloomfield kindly allows me to include some additional records published by him in the Transactions of the Norfolk and Nonoich Naturalists' Society (vol. vii. pp. 227, 427) and in the Victoria County History ; these are indicated by the initials E. N. B. A few records are quoted from the Census Catalogue of British Mosses. These are given on the authority of the late Prof. Barker, but owing to his death information as to the exact localities whence they were obtained has been lost. All the plants of my own gathering have been submitted for verification and correction to Mr. Wm. Ingham ; I have also received much valuable help from Mr. Bloomfield and Messrs. W. H. Burrell, H. N. Dixon, and W. E. Nicholson, to all of whom I tender my thanks. Names of mosses hitherto recorded for Suffolk are preceded by an asterisk. Sphagnum acutifolium Ehrh. var. suhnitens Dixon. 25 and 26. Eedgrave Fen, east and west. (The boundary line between the two vice-counties divides the fen into two nearly equal parts.) — S. fimhriatum Wils. 26. West Suffolk, Census Catalogue. Polytrichum strictum Banks. 25. Herringfieet {Turner, 1806), E. N. B. — P. gracile Dicks. 25. Mendlesham. THE MOSS FLORA OP SUFFOLK 73 ■'Dicranella Schreberi Schimp. 25. Mendlesham. ■''Fissidens viridulus Wahl. 25. Mendlesham. — *Var. Lylei Wils. 25. Mendlesham and Old Newton. — F. incurvtis Stark, 25. Mendlesham; Gipping. — F. bryoides Hedw. 25. Mendlesham ; Brockford. — *F. decipiens De Not. 25. Stuston Common. Bhacomitrimn canescens Brid. var. ericoides B. & S. 25 or 26. Suffolk (Eagle), E. N. B. Phascum cuspidatum Schreb. var. piliferum Hook. & Tayl., and *var. curvisetum Nees & Hornsch. 25. Mendlesham. Pottia crinita Wils. 26. West Suffolk, Cenaus Catalogue. — P. Starkeana C. & M. 25. Belton (Borrer), E. N. B. ; Greeting. 26. Bury {Bunbury), E. N. B. Tortula pusilla Mitt. 25. Mendlesham. — T. Icevipila Schwaeg. var. IcBvipilaformis Limpr. 25. Grundisburgh (W. B. Sherrin), E. N. B. — T. ruraliformis Dixon. 25 and 26. East and West Suffolk, Census Catalogue. ■'Barbula lurida Lindb. var. intermedia Euthe (nearly approach- ing B. cordata Dixon). 25. Needham Market. — B. tophacea Mitt. 25. Mendlesham ; Playford. — B. cylindrica Schimp. 26. Drink- stone churchyard. — B. vinealis Brid. 25. Mendlesham. — B. sinuosa Braithw. 25. Sweffling, E. N. B.; Mendlesham ; Need- ham Market. 26. Great Finborough. — -'B. Hornschuchiana Schultz. 25. Mendlesham. — B. convoluta Hedw. 25. Mendles- ham ; Playford. OrthotricJmm cupulatu7n Hoffm. 26. Sweffling, E. N. B. 26. Great Finborough. — 0. tenellum Bruch. 25. Finningham. — 0. pulchellum Smith. 25. East Suffolk, Census Catalogue. Mr. Bloomfield assures me that he has it on good authority that Braithwaite's record of this plant from Burgh Castle was erroneous. Physcomitrella patens B. & S. 25. Mendlesham, Philonotis fontana Brid. var. falcata Brid. 26. Tuddenham, Skepper. Eecorded as P. calcarea in Journ. Bot, 1885, 286, but corrected by Mr. Bloomfield in Trans. Norf, & Norw. Nat. Soc. vii. 427. Leptobryum pyriforme Wils. 25. Ditches at Mendlesham and Playford. Bryum pendulum Schimp. 25. East Suffolk, Census Catalogiie. — B. pseudo-triquetrum Schwaeg. 25. East Suffolk, Census Cata- logue. — *J5. erythrocarpum Schwaeg, 25. Mendlesham. — *£. miirale Wils, 25. Mendlesham. Mnmm affine Bland. 25. Stuston Common ; Nacton ; Greeting, Cryphcsa heteromalla Mohr, 25, Trees about Yarmouth, Botanists' Guide. Pterogonium gracile Swartz, 26. Icklingham (Eagle), E. N. B. in Victoria County History. Thuidium abietinum B, & S. 25, Souston Common, — T. hystricosum Mitt. 26. Barton Mills (Borrer, BraitJmaite), E. N. B. — *r. recognitum Lindb. 25. Stuston Common and Greeting. Journal of Botany. — Vol. 52. [March, 1914.] g 74 THE JOURNAL OF BOTANY ■'Climacium dendroides Web. & Mohr var. fluitans Hiib. 25. Stuston Common. Brachythecium rivulare B. & S. 25. Playford. Eurhynchium Sivartzii Hobk. 25. Mendlesham. Aviblystegium varkim Lindb. 25. Mendlesham ; Old Newton ; Needham Market. — -''A. irriguum B. & S. 25. Mendlesham. ■'Hypmim riparmm L. var. subsecimduvi Schimp. 25. Mendles- ham.—iZ". elodes Spruce. 25. Eedrave Fen, east. — "'H. stellatum Schreb. var. 23^ote7is7tm Eohl. 25. Mendlesham. — "H. aduncum Hedw. var. falcatmn Hedw. 25. Mendlesham. — "=Var. gracilescens Schimp. and var. j^olycarjwn Bland. 26. Knettishall. — '"'Var. intermedium SGhixmp. 25. Eedgrave Fen, east; Stuston Common. — "'''-Var. patermtm Sanio. 25. Mendlesham ; Stuston Common. — ■'H. Sendtneri Schimp. 25. Stuston Common. — H. Wilsoni Schimp. var. hamatum Lindb. 26. Tuddenham [Skepper), E. N. B. — H. lycopodioides Schwaeg. 25. East Suffolk, Censtis Catalogue. — ■■H. revolvens Swartz. 25. Eedgrave Fen, east. — H. cordifolium Hedw. 25. Bogs on Bradwell and Belton Commons, Botanists' G^iide. — H. giganteum Schimp. 25. Eedgrave Fen, east. A NEW ANNONA FEOM JAMAICA. By W. Fawcett, B.Sc, & A. B. Eendle, F.E.S. Annona prsetermissa, sp. nov. ^r6or parva, ramulis novellis puberulis mox glabris. Folia elliptica vel ovato-elliptica, apice subacuminata, basi late cuueata, supra basim versus atque in costa media subinde pubescentia cetera glabra, subtus subtiliter adpresse pubescentia, supra costis planis et venis subobsoletis infra costis prominentibus et venis planis ; petioli pubescentes. Pedunculus biflorus, tomentosus ; pedicelli supra medium bracteolati, tomen- tosi. Alahastra acuminate conica. Sepala late deltoidea, tomen- tosa. Petala tria, oblonga, extra tomentosa, intus subcarinata. Fruct2cs globosus, areolatus, tuberculatus, tuberculis apice hamatis. Type in Herb. Jam. A small tree, about 15 ft. high. Leaves 12-18 cm. 1., 4'5-8-5 cm. br. ; petioles 13-16 mm. 1. Peduncle very short to 1 cm. 1. ; pedicels very short to 1-4 cm. 1. Sepals 2-5-3 mm. 1. Petals 2'5 cm. 1., -5 cm. br. Stamens 2 mm. 1, ; anther about 1*5 mm. 1. Fruit about 6 cm. in diam. Seeds 17 mm. 1., 10 mm. br. Hab. — Craig Hill, near Gordon Town, June, 1902, Faiocett ! Near A.jamaicensis Sprague, but easily distinguishable by the longer, oblong petals, the conical, not ovoid, buds, the areolate fruit and larger seeds. 75 EOBEET JACOB GOEDON. (1741-1795.) By James Britten, F.L.S. The Biographical Index of British and Irish Botanists con- tains the following entry : — "Gordon (fl. 1774-79). Colonel. Travelled in Africa, 1774, and, with Paterson, 1777-79. Discovered and drew many StapelicB. Masson, ' Stapelia,' pref. viii. ; Journ. Bot. 1884, 145. Stapelia Gordoni Mass. This summary, which represented all that was known at the time of compilation, may now be considerably amplified ; as a conse- quence of this, Gordon's name will disappear from the next edition of our book, in which it is clear he has no claim to be included, but it may be worth while bringing together such information as exists about a remarkable man. In the sale of the Stafford Library at Sotheby's in November last was included a collection of 400 watercolour drawings made by Gordon in South Africa from 1777 to 1790 ; the drawings — which included natives, quadrupeds, birds, reptiles, and plants as well as plans and views of the district round the Cape and the Orange Eiver — although somewhat crudely executed, are evidently extremely accurate, and are furnished with descriptions in Dutch : the views and plans, many of which are very large, are in two volumes atlas elephant folio, the drawings in four volumes elephant folio. The collection was purchased by Messrs. Maggs Brothers, of 109 Strand, for £690, and was promptly placed on the market by them at almost double the price — £1250. A special catalogue ("Supplement to Catalogue 316") was issued by them which contains reproductions of two of the maps, of the drawing of a Hottentot man, woman, and child, of two native groups, of a giraffe and another animal, and lists of the animals and plants figured — of the latter in somewhat unfamiliar form — e.g. " Orthnathogicum " for Ornithogalwn. The early history of the collection, however, is given in a letter to Banks from Philip Gidley King (1758-1808), who at the time of writing was Lieutenant-Governor of Norfolk Island but was then in London. A copy of this is preserved in the Banksian Correspondence, and may be worth transcription : — " London, May 27th, 1797. " Sir, " Agreeably to your wish, I have informed myself more fully respecting the Papers of the late Colonel Gordon, brought to this Country by his widow. " The Charts, &c., are contained in two Boxes (which I saw inspected yesterday at the Custom house). The largest Box contains, as Mrs. Gordon informs me, a general Chart, smaller Charts and Views of the Interior parts of Africa seen and visited by her late husband, in all about ninety-five, with a Manuscript a 2 76 THE JOURNAL OP BOTANY account wrote in Dutch. There are also a few bundles of family Papers. The second Box contains a very full and large Book, in which are arranged upwards of 400 drawings of Natural History, appropriate to the Charts and Views. " The Charts and Natural History Mrs. Gordon informs me were all designed by her own husband, who drew every outline, and had them finished under his own eye. As her wish is to have these Charts, &c., inspected by such persons as may be deemed adequate to judge of their consequence to this kingdon, she desires me to request in her name the indulgence of their being permitted to be withdrawn from the Custom house, where they are now lodged, without being subject to the duty. " I beg leave to apologize for the part I have taken in this business, to which I am alone prompted by the respect I bear to the memory of her deceased husband, and her situation as a stranger in this Country, from whence it is her intention to depart with her family for her native Country, Switzerland, the instant her business is finished. — I have the honour to be most respectfully, " Sir, Your most obedient, " humble servant, " Philip Gidley King." From this it is clear that Banks knew of the collection and had some thoughts of obtaining it. Whether it actually reached him I do not know ; but it may be noted that in the Banksian collection of Masson's drawings (see Journ. Bot. 1884, 144; 1885, 227) are two — Hoodia Gordoni and Pachypodmm naviaqiianum — which are labelled " Webber " in Dryander's hand, the latter being noted by him as "copied from a drawing of Captain Gordon's at the Cape of Good Hope"; it may however well be that this information was supplied to Dryander by Masson. This drawing is reproduced in Lieut. William Paterson's Narrative of Four Jour- neys into the Country of the Hottentots and Cajfraria (1789) : it seems likely that the other figures of plants in the volume are from the drawings of Gordon, who accompanied Paterson on his jour- neys in 1777 and 1779, and to whom the latter frequently refers. Masson (StapelicB, viii.) mentions both Gordon and Paterson — who in his book always spells his name " Mason " — as having "discovered some very remarkable species of Stapelia" ; his description and figure of S. {Hoodia) Gordoni are taken from the drawing referred to above. According to the Maggs Catalogue, " Colonel Eobert Jacob Gordon was a Dutchman of Scottish extraction, born in Guelder- land in 1741. He was in command of the Dutch forces at the Cape, and it is said that when the English took the Colony in 1795 he shot himself in chagrin at the failure of his resistance to our arms." Paterson (L c. 113), under date August 1779, gives an account of his naming of the Orange Eiver in honour of the Prince of Orange, which however [l. c. 61) he seems to have done previously in 1777. Incidentally, it would appear that it was SHORT NOTES 77 indirectly to Gordon that Australia is indebted for its great wool industry, as liis widow in 1795 sold to the commanders of the Siipply and BoUance sloops of war some merino sheep which were taken to Australia to Captain John Macarthur, who was then experimenting in wool production in New South Wales. A warm appreciation of Gordon will be found in John White's Journal of a Voyage to Ncio South Wales (1790), p. 90. White visited Gordon at the Cape in 1787; his garden, he says, displays " not only the taste and ingenuity of the gardener, but the skill and knowledge of the botanist." " The Colonel is a man of science of an active and well-cultivated genius, who appropriates those hours he can spare from his military duties (in which he is said to excel) to a perusal of the book of nature and researches after useful knowledge." It was his intention " to publish the observations and remarks which have been the result of his researches," and it is to be regretted that this intention was never carried out, as we may share White's conviction that Gordon had " made himself better acquainted with the subject, and pene- trated farther into the interior parts, than any traveller or naturalist that [had then] visited the Cape." It is to be hoped that the collection of Gordon's drawings may find its way into the possession of someone who will see that lists of the species collected by him are drawn up by competent hands. SHOBT NOTES. Ptilota plumosa Ag. in Devon (p. 35). — Mr. A. D. Cotton, writing of Ptilota plumosa and two other algae, says that " none of them are found in any part of the English Channel. ... it was not easy to understand why these plants should not extend by way of the Welsh coast to Devon and Cornwall " ; and in a foot- note to p. 39 says that a fragment of a frond of the Ptilota inscribed " Ilfracombe, E. T." occurs in Prof. Phillips's collection, but " as the plant does not occur in this well- worked region, the locahty given must be regarded as erroneous." In July, 1907, I found, about four miles east of Ilfracombe, a plant I sent to Mr. E. M. Holmes, which he returned to me labelled, " It certainly is Ptilota plumosa," with drawings made by him on the back to show its structure as compared with that of P. elegans. — C. E. Larter. A Correction (p. 43). — The interviewer from the Morning Post who visited me on Jan. 13th, in writing out his notes has mixed up two entirely distinct statements. He asked me. How many plants are there in Tropical Africa? I told him it was impossible to estimate the number, because many districts had not been explored yet, and gave him an illustration of this — Mr. and Mrs. Amaury Talbot's collection from Southern Nigeria recently worked up and published by the botanists of the Natural History Museum. This contained 600 numbers, of which 150 78 THE JOURNAL OF BOTANY proved to be new. Then we spoke of the plants of the whole world ; I gave him 200,000 species as a rough estimate. When his report was published I was startled to find that it stated Mr. and Mrs. Talbot's Nigerian collections numbered 200,000 instead of 600 species ! — J. G. Baker. REVIEWS. The Story of Plant Life in the British Isles : Types of the Common Natural Orders. Introductory volume by A. E. Horwood. Illustrated with 73 [figures from] photographs. 8vo, cloth, pp. xiv, 243. London : J. & A. Churchill. Price 6s. 6d. net. It is forty-five years since Sir Joseph Hooker pointed out, in the preface to his Student's Flora, the need for a companion volume to that work which should summarize "those physiological and morphological observations on British plants which have of late given so great an impulse and zest to botanical pursuits," and held out a prospect that he might at some future time be able to undertake the task. Long as his life was, this hope was never realized, and the need for such a work is far greater than ever. Indeed, it may well be doubted whether any single volume of reasonable extent could adequately present even a summary of such observations, to which would have to be added some con- sideration of the investigations grouped under the name of ecology. The statement in the preface to Mr. Horwood's book that he had " endeavoured to give briefly a connected account of the essential phases of the life-history " in the case of the plants selected for description led us to hope that the scheme proposed by Hooker had at least been attempted ; and the announcement that his " method of description is an advance upon previous works of the kind " induced pleasurable anticipations. We regret that in neither respect have these anticipations been fulfilled. We cannot see that his book as a whole differs to its advantage from many of those already on the market ; from the literary point of view it is indeed distinctly below them, for Mr. Horwood's style is involved, and it is not always easy to determine what he means. This criticism may sound harsh, but we do not think anyone w^ho will read the first five paragraphs of the Introduction will consider it too severe; we will quote only the fifth. Having told us that " physical surroundings play a great part in the shaping of species, apart from their diversity and the fact that these diversities are correlated with plant distribution and plant variation," Mr. Hor- wood continues : — " But this is not all, for we learn from the character of the surroundings its requirements as regards light, heat, moisture, altitude, soil, &c., and the manner in which the plant occurs, either in small communities, large ones or otherwise helps us to obtain a much broader and more intelligent view of the vegetation of a district, or its physiognomy on a large scale, which in turn PLANT LIFE IN THE BRITISH ISLES 79 reveals to us the bases of scenery and landscape. So that here the painter or the poet may join in the study of botany from a really vital standpoint." It is to be regretted that the school- mistress whose help in revising the proof-sheets Mr. Horwood acknowledges did not add to her " many helpful suggestions " one as to the need for clearness of expression and another as to the principles of punctuation. Leaving the Introduction, in which is much that might be criticized, we come to the descriptive portion. Mr. Horwood's plan is to select for description "common types" of "the more widely distributed and more familiar orders." We are unable to discover on what principle his selection has been made :_ thus, in Monocotyledons, Melanthacece, with one representative, is included, while Liliacece, and AmaryllidacecB, which include the wild Hyacinth and the Daffodil, are omitted ; in GluviacecB we have some account of Gyperacece and a description of Eriophonm, but the Grasses are entirely absent, and this from a work which purports to be a " handbook of the common Natural Orders " 1 " The rarer representatives of the orders will be dealt with in a forthcoming work" — a statement which seems to mean (see p. 219) that the orders not included in this volume will appear there : anyway, it is not easy to see how the Daffodil, the wild Hyacinth, and the whole of the Grasses can be included under this head. Nor is "the beginner" greatly helped by being told that " reference can easily be made to more comprehensive works." What he wants is one book, and there is no reason why a volume of this size should not suffice his requirements. Turning to the " life-history " of the plants, which it is the main object of the book to present, it is fair to say that Mr. Hor- wood gives a good deal of useful information as to the habits_ of the plants described, and details connected with fertihzation receive more attention than is usual in popular books. In this respect the author's hope may claim to have been to a certain extent reaUzed, but his treatment leaves much to be desired. We find no account of the seedling state of any species : nor, to take a single example, is any reference made to so common and striking an occurrence as the propagation of Cardamine pratensis by means of its leaflets, or to the curious and almost equally frequent prohferation of its flowers. Omissions of this kind are serious in a book the object of which is "to bring the student iiito the field [author's italics] to study " ; and we are bound to say that we do not find in his descriptions much indication of original observation. The book suffers throughout from want of arrangement, which leads to useless repetition. To take an example, the Holly is described both under " the Holly group" (pp. 78-80) and at length under its special heading (pp. 80-82) ; if the two descriptions had been combined, and the matter rearranged, at least a page would have been saved. Its chief defect, however, is in the prominence of matter in no way relevant to " life-history," although painfully familiar to readers of "popular" books. Still keeping to the account of the Holly, we find the following : — 80 THE JOURNAL OP BOTANY " In Northumberland holly leaves were used for divining. It was planted near houses to ward off lightning, as early as the days of Pliny. Because it resembles the word ' holy ' it was reputed to be inimical to witches. Holly wreathes [sic] were employed in Eoman times at weddings. People used to cure their chilblains by threshing [sic] them with holly leaves. The bark has been used in place of cinchona, being astringent " (p. 82). It is difficult to imagine that unauthenticated scraps of this kind — even if accurate, which we do not think is always the case — can be of the slightest value to any serious student. The same may be said of the scraps of verse — Shakespeare is misquoted on p. 55 and Samuel Lover on p. 96 ; the dedications — there is no authority for saying that Caltha " is called Marigold because it was dedicated in mediaeval times to the Vii'gin Mary " ; snippets from Gerard, "Baldur" (p. 141), and the like; and the references to foreign species. A further example of padding is found in the space devoted to the popular names of the species, which have been appropriated wholesale from the Dictiojiary of English Plant-names. Mr. Horwood has not even taken the trouble to consult the body of that work: he has simply "lifted" from the index, in which all the English names are conveniently placed under their Latin equivalent, such portions as suited his purpose. He has not even taken the trouble to do this correctly ; thus in taking the twenty- three names of the Ash (p. 158) he misprints " Urchin, Wood- broney " as "Urchin Wood, Croney" ; and tells us that "Esh " — the north-country variant of Ash — "means to flog, the twig of an ash being used for the purpose " ! This astounding derivation, worthy to stand beside Dr. Brewer's "Coltsfoot — cold's food, i.e. food for colds and coughs" — is at any rate, in common with others (see "Bow Thistle," p. 115) equally ridiculous, Mr. Horwood'sown; the Dialect Dictionary gives no such use of the word. Nor has he even appropriated intelligently, for he includes words obsolete and of doubtful application; thus he gives as " common names of the Violet " several to which in the Dictionary a " ? " is attached. Under Tilia we have this amusing note : " Though Pliny gave the name ' Tilia,' there are some old vernacular names that might equally have given origin to it, such as Telle, Til," &c. — names which anyone but Mr. Horwood would have seen are derived from Tilia. Perhaps, considering the numerous misprints and the unintelligent way in which the names are printed, it is as weU that Mr. Horwood should not have acknowledged the source of his information, even by placing it in the very inadequate biblio- graphy (p. 222) ; nevertheless this wholesale appropriation of other men's work calls for explanation either from the author or from his pubhshers, whose attention we call to the fact. The numerous figures from photographs which accompany the descriptions are very unequal : some — c. g. Violet, Sloe, Angelica, and Coltsfoot — are good, others — e. g. Broom, Meadow Cranesbill, Groundsel — the reverse. The indications of size are sometimes misleading — e. g. the flowers of Stitchwort and Oxeye Daisy do MILDEWS, RUSTS, AND SMUTS 81 not appear to us " enlarged." As a rule the attempt to show the whole or a considerable portion of the plants leads to indistinct- ness— e. g. the Gean, Tufted Vetch, Creeping Buttercup, and many others ; the Cuckoo Flower shows only white flowers standing out of a black background. This method of illustration has its advantages, but unless it is very well carried out, these are not always apparent. It remains to be said that the volume Would be improved by more careful reading ; we have already referred to the need of revision as to composition and punctuation, and to the numerous misprints in the names extracted from the Dictionary of English Plant-names, but others occur — e.g. Dioscoreacece is spelt "Diosco- raceae " (pp. x, 198 (twice), 209, 244) and " Dioscoriace^ " (p. 222), but never correctly; " Anagrgecum" (p. 201), "Tofeildia " (p. 206, thrice), " Brittanicse " (p. 224). We should not have noticed the book at such length had it not been for the author's somewhat pretentious estimate of its importance and for the fact that it is the first of three volumes, the remaining two of which are "to be published very shortly." It would have been easy to have extended our criticism ; but sufficient has been said to show that Mr. Horwood would do well, before sending these to press, to submit them to a friend who has a blue pencil and is not afraid to use it. We would also suggest that, should he wish to borrow extensively from works already in existence, it would be courteous to obtain permission to do so, or at least to acknowledge the source of his information. James Beitten. Three Books on Fungi. 1. Mildetus, Busts, and Smuts: a Synopsis of the Families Perono- sporacecB, Erysiyhacea, UrcdinacecB, and Ustilaginacea. By George Massee, assisted by Ivy Massee. Pp. 229 ; 5 plates, 1 coloured. Dulau & Co., Ltd. 1913. 7s. 6d. net. 2. The Fungi ivhich cause Plant Disease. By F. L. Stevens, Ph.D. Pp. ix and 754 ; 449 figs. New York : The Macmillan Company. 1913. 17s. net. 3. The Diseases of Tropical Plants. By Melville Thurston Cook, Ph.D. Pp. xi and 317 ; 85 figs. Macmillan & Co., Limited, St. Martin's Street, London. 1913. 8s. 6d. net. 1. Mr. Massee has added another to his already long list of fungus text-books. The groups included are, with the exception of Perisporiacese, indicated in the subtitle, and a "chapter" is devoted to each. An account is given of the interesting points in the life-history of each family, followed by a description of the essential characters. A useful generic key is added and a note to each genus. The genera and species are then described. In the case of Peronosporaceae a key is given to the species of each genus. An innovation is the introduction into these keys of fungi not yet found in this country but which are liable to be met with as the host-plants are present. In the rusts, descriptions are given of European species of Puccinia and Uroinyces parasitic on native 82 THE JOURNAL OF BOTANY British plants, bub not recorded as occurring in Britain. The idea is excellent, but it is rather disconcerting to find that in both cases many well-authenticated British records have not been accepted. The treatment of the Peronosporaceae and Erysiphaceee calls for no special comment, save that what is said of Phytophthora erythroseptica shows that Pethybridge's account of the life-history of the fungus has not been carefully read, and the comments indicate a want of knowledge of recent work on the genus. The treatment of the Uredinales seems far from satisfactory. Certain species of ^cidiwn and Ureclo are placed in an appendix, though in some cases, e. g. ^cidnmi leucospermum, the alternate stage is well known on the Continent, and ought certainly to have been mentioned. To accept most of the biological species of Puccinia and then to give only three species of Melampsora seems illogical. It is interesting to find that Phragmidium Fragariastri is made to include P. Poterii, P. Sanguisorbce, P. Potentillce, and P. Tor- mentillce on the ground of morphological transitions. Truly, P. TormentillcB seems at present to be undergoing many vicissi- tudes (c/. p. 53). The author also considers that Hemileia amoricana is our only British species. There are many misprints. Mistakes are also far from infrequent. " The only British species " of Endopihyllum (p. 68) becomes the usual two on p. 93. Bifi'en (p. 129) is given credit for experiments pei'foi-med by Tranzschel and confirmed by Brooks. The collaborator of Pethybridge in the investigation of the potato disease becomes "Murray" instead of the more appropriate "Murphy." The " ascigerous condition" of Calyptospora Gosppertiana is given in plate iii. The genus Chrysomyxa seems to be absent altogether, except in one of the plates, whilst Pucciniastrum is apparently represented by one species of Uredo placed in the appendix. There is little, if any- thing, new in the treatment of the Ustilaginales. In three places are statements with regard to the lack of books dealing with the microscopic fungi here considered. "It is now nearly half a century since the last British book on fungi, includ- ing the rusts and mildews, was published." This presumably was Cooke's Handbook. The statement is misleading. The fact is that it is now customary to monograph groups separately, and not to consider collectively five families which have practically nothing in common. Perisporiaceae (two genera, four species) is the only family which has not been treated by a British author since Cooke's book, which included the whole of the fungi. The book contains an index of genera and species and another of host-plants. Both appear to be very good. The plates give clear line drawings, but their usefulness is questionable ; a few text-figures would have been of greater aid to students. The book is of a size that makes it possible to be carried in the pocket. If more care had been taken with it, we imagine that the book, which is strongly bound and well printed, would have been the field companion of most British mycologists ; even as it is, it will be found the most convenient for work in the field. J. Eamsbottom. FUNGI WHICH CAUSE TLANT DISEASE 83 2. If any justification were needed for adding to the growing library of works on plant diseases, it would be found in the economic importance of the subject. The cultivation of plants involves the crowding together of species which entails the danger of epidemic attack, so that some comparatively innocuous and negligible fungus becomes by opportunity a destructive parasite. Thus new attacks by fungi or by animals are constantly being discovered, and the life-history of the organism causing the disease must be traced before effective remedial measures can be applied. The books before us are both by American writers well-known for their work on the diseases of plants, and though they are more or less restricted in scope they are of interest and value to all students. The volume by Dr. Stevens is, he tells us, " intended to intro- duce to the student the more important cryptogamic parasites affecting economic plants in the United States," but as fungi, and especially those that cause diseases of plants, are largely cosmo- politan, the book is fitted to be of world-wide service. It often happens that some part of a plant already injured becomes covered with fungal growths, and it is extremely important to know whether such fungi are of parasitic or saprophytic habit. Dr. Stevens has gone over the whole field from Myxomycetes to Fungi imperfecti, selecting those fungi that are known to have caused trouble. He gives synoptic tables and diagnoses of families and genera that are either proved or suspect as the cause of disease, and lists the species that are fatal ; in the case of American species he supplies full descriptions. The Fungi impcrfecti are classified and described as such, but where the full life-history of these forms has been worked out, references are given to the perfect forms. The book is well illustrated, and is provided with copious bibliographies, a glossary of the terms employed in the text, and an index of hosts and parasites. It can be recommended with confidence to all students of this very extensive and difficult branch of botany. 3. Dr. Cook deals only with the diseases of tropical plants. The demand for the vegetable products of the tropics, such as cotton, rubber, cocoa, &c., has increased enormously, and the book has been written with a view to help the planters in their struggle with new and adverse conditions. The first few chapters give in order a survey of plant life ; the nature and symptoms of disease ; the structure and function of plants ; the classification of fungi ; and an account of various causes of plant disease. Other chapters are devoted to a short account of slime-moulds, phanerogamic parasites, bacteria, insects, worms, and, lastly, functional or physiological diseases. Most of the diseases are due to fungi, and a discussion of these occupies the larger part of the earlier chapters. The second and more extended part of the work deals with the host-plants in due order and the maladies that attack them, starting with corn and rice, and winding up with forest and ornamental trees. Preventative and curative agencies are also 84 THE JOURNAL OF BOTANY discussed. The subject is widely conceived, and, in the limited space, fairly well carried out. It is, however, the fault of many similar books that, in the attempt to tell everything, much is left unexplained. Thus we find that " damping-off " of seedling sugar- canes is caused by three different fungi — Pythium Deharyanuvi, which is fully described in the introductory chapters, Bhizoctonia, and Glomerella ; there is no indication, however, as to the appear- ance or the affinities of the two latter — they are mere names. The same inequality of treatment is meted out to many other genera and species. Fungi imperfccti receive scant attention, though they are responsible for many leaf-diseases. Many of them are the imper- fect or development stage of some Ascomycete or Basidiomycete, and in not a few instances the full life-history is well known ; yet Dr. Cook writes, "The Fungi ivqjcrfecti are so called because we do not understand their life-history and development." The illustrations are abundant and instructive, and the book will doubtless be of great value to the agriculturist in the tropics. A bibliography is appended of the literature dealing with culti- vated tropical plants. « t c< ^ ^ A. LoRRAiN Smith. The Flora of South Africa. By Kudolf Marloth. Vol. i. 4to, with 36 coloured and 30 monochrome plates. London : Wm. Wesley &■ Son. Subscription price for the volume £2 2s. By his collections of South African plants and his published memoirs Dr. Marloth has won a deservedly high reputation among students of South African botany ; it was fitting, there- fore, that he should have been entrusted with the task of pro- ducing a series of four well-illustrated volumes designed to do justice to the varied and wonderful flora with which he is so well acquainted. In these days of research in all directions, a mere enumeration of genera and species is obviously insufficient for the purpose, since such an enumeration, while indispenable for deter- mining the position and affinities of a plant, leaves the whole romance of its history untouched. Enquiry is much concerned nowadays with the relations of plants to their environment, with the causes underlying the phenomena of their distribution, with details of their structure and chemical constitution, and with the ways in which they are or may be rendered serviceable to the ever-growing wants of mankind. It is from all these points of view that Dr. Marloth has endeavoured to treat the subject. He first gives the characters of a family and then a key to the genera included in it ; this is followed by an account, often at con- siderable length, of species selected as possessing peculiar interest in one or more of the ways above-mentioned. When it is added that the families are sufficiently and often lavishly illustrated by means of coloured and monochrome plates, reinforced by a num- ber of figures in the text, it will be seen that high things have been aimed at, and so far at least as the volume before us permits of a judgment, we have no hesitation in saying, with unqualified success. The present volume commences with the Thallophyta, treated FLORA OP SOUTH AFRICA 85 somewhat shortly but at sufficient length for the student. The ArchegoniatcB follow, the synopsis of genera and illustrations of each of the families being particularly good. The space devoted to the Gymnosperms has been wisely utilised, and full justice is done to the Cycads and Welwitschia. These items disposed of, way has been made for the Dicotyledones, which occupy the bulk of the volume ; of the 141 families of these plants recognised by the author 31 are treated here, including all the Monochlamydea and of the Dialypetalm the Eanales and the Rhoeadales. The system followed is mainly that of Engler's Synopsis, a system not without its faults, but on the whole an improvement on the linear arrangements previously in vogue, though to old-fashioned folk it will be somewhat disconcerting to find, for instance, Buttercups and Water Lilies in close juxtaposition to Laurels and GeratophyllacecB. But the author has endeavoured to bring his work, so far as is under the circumstances possible, into agree- ment with that of his predecessors, references to the Flora Capensis being numerous, while Harvey's Genera of South African Plants is cited throughout. The treatment and illustrations of the parasitic families Lorantliacece, HydnoracecB, BalanopJioracece and Bafflesiacece are worthy of special mention ; also of that family, so important at the Cape, the Proteacea, as of the gall- flowers of the figs. Best of all seems to us the notice given to Mesembriantheniuin, to which genus 26 figures in all are devoted ; the person who can read the fascinating story of the " window- leaves" of certain species of this genus without a thrill of pleasure must, indeed, be without a spark of love for Natural History. We know of nothing like the present work upon the flora of any of the overseas Dominions, and the Homeland itself, it must be confessed, is in the same predicament. To Dr. Marloth should therefore be given all the credit owing to, though, unfortunately, not always achieved by, those who act as pioneers in the advance of science. He has had assistance from many quarters as, for instance, in the preparation of the drawings for the plates, and Professors Diels and Brotherus and Mr. Sim have taken a hand in the construction of the synopses of genera, while Mr. W. T. Saxton's help has been available for the Thallophyta. The incep- tion of the undertaking is due to Lady Phillips, without whose munificent support it could not have been realised, and who has, in consequence, earned a debt of gratitude from all those who, whether experts or students or travellers, will profit by this truly remarkable production. The printing, by the Cambridge Uni- versity Press, is a fine example of the workmanship turned out at that well-known establishment. o lyj- A Naturalist in Western China. By Ernest H. Wilson. 2 vols, with 101 plates. Methuen. Price 305. net. There is a threadbare proverb about having too much of a good thing, the converse of which would, one imagines, find a more responsive echo in the experience of mankind at large ; certainly it was our feeling when we reached the last page of 86 THE JOURNAL OF BOTANY Mr. Wilson's second volume. We could have wished to know more of his adventures during the eleven years he spent in China — at first on behalf of Messrs. Veitch, afterwards under the auspices of the Harvard University's Arnold Arboretum. As his map shows, he must have toiled along many a breakneck track, escaped perils, witnessed remarkable scenes of which he tells us nothing. A too modest estimate of his narrative powers, above all the desire, as far as possible, not to re-traverse a field where he has had several forerunners, are the considerations that have led him to restrict his personal story. The provinces of Hupeh, from Ichang westward, and of Szechuan, including in the latter the little known Chino-Tibetan marches, is the region dealt with in these volumes. The difficulties of travel in this part of the Middle Kingdom must be enormous, entailing as it does carriage of bed and bedding, food and cooking utensils, over some of the most villainous tracks trodden by foot of man ; nor should that sign of respectability, a chair, be omitted, nor, significant item, a store of insect powder ! So pro- vided Mr. Wilson fared cheerily onward, ever on the q2ii vive for botanical spoil, while his companion, Mr. Walter Zappy, busied himself with the rich fauna of the country. We pass with him in spirit through parts of Hupeh hitherto unvisited by Europeans. He takes us across Eichthofen's Red Basin of Szechuan, with its wonderful irrigation canals, and so into the mountainous western borderland. For him the mountains had a special call ; it was on them he expected to make his principal finds, and, as many a herbarium and the Plantce, WilsoniancB, now in course of publica- tion, testify, the harvest he reaped was indeed a splendid one. And it was at the very threshold of this marvellous region, the home of the richest temperate flora of the globe, that Maries turned in 1879 under the impression that Eobert Fortune had already secured virtually all that was worth collecting in China ! Sixty-eight thousand specimens, comprising about five thousand species and seeds of over fifteen hundred plants, were the si^olia opima Mr. Wilson brought home. How many of these species are new to science cannot at present be stated ; but we do know that the labours of recent travellers, such as Mr. Pratt, Professor Henry of Cambridge (not Dublin, Mr. Wilson !), and the French col- lectors, together with those now under notice, have revolutionised even the best-informed botanists' notions of the Chinese flora. Mr. Wilson's predecessors, however, have given us no record like the present volumes, with their many illustrations of scenery, their exciting descriptions — at one place the author had to pass through masses of Cypripediums so dense as to render crushing the flowers unavoidable ; their carefully worked-out zones of mountain vege- tation and valuable information for botanist, horticulturist and forester. Several excellent wild fruits were found, which, by grafting, will be the starting point, it is to be hoped, for new and improved varieties. We agree with Mr. Wilson in thinking that the value of certain Chinese reputed medicines should be enquired into ; true, some of the remedies now in use are as absurd as BOOK-NOTES, NEWS, ETC. 87 those European mediaeval compounds which cause us a shudder and a smile ; but his own experience seems to show that, though much of the pharmacopoeia is fantastic and worthless, there is a residuum worthy of serious testing. On the whole, Mr. Wilson was favourably impressed with the people of the country ; but he is de- servedly severe on their want of thrift in clearing away the forests, thus leaving a menacing problem for the near future ; also on the poor quality of much of their fruit, the result of slovenly cultivation. Charles Lamb, dilating on the catholicity of his taste, tells us he banished from his shelves only such books as " no gentleman's library should be without." This attitude should, and doubtless would, have caused the great humourist to look askance on A Naturalist in Western CJiina, for it answers fully the qualifica- tion entailing such banishment, provided, of course, the "gentle- man " in question has a taste for natural history ; not botany alone, since zoologists, geologists, ethnologists, and sportsmen as well as traders will all find something in these pages to claim their attention. But the appeal in chief is to the botanist and horticulturist, an appeal emphasized by Professor Sargent's essay comparing the forest floras of China and Eastern North America. We trust these handsome volumes, the outcome of indomitable pluck and patience, will meet with the success they so richly merit. S. M. BOOK-NOTES, NEWS, dc. The second volume (first in appearance) of the Cambridge British Flora is announced to appear in March. It will contain the orders Salicaceae, Myricacese, Fagacese, Corylaceoe, Betulacese (by the Eev. E. S. Marshall), Ulmacese, Cannabaceae, Urticacese, Santalaceae, Loranthaceae, Aristolochiacese, Polygonaceae, Amaran- taceae, and will be mainly the work of the editor, Dr. Moss, who in Chenopodiaceae will have the assistance of Mr. A. J. Willmott in A triplex and of Mr. E. J. Salisbury in Salicornia, the genus Salsola being treated by Mr. C. E. Salmon. The order Aizoaceae is included in virtue of the Mesemhryanthemuvi naturalized in Cornwall. The prospectus containing information as to price, format, &c., of the work, with a specimen plate, may be obtained from the Cambridge University Press. Of the Eev. E. F. Linton's Supplement to the " Set of British Salices," to which reference was made in this Journal for 1913 (p. 232), two fascicles have been issued — the first in June last, the second at the end of last year. From the information issued with the second fascicle we learn that " The present venture has been started on a co-operative basis ; much material has already been sent in, and more has been offered or promised, by Dr. G. Fogerty, Messrs. E. A. Phillips and A. E. Bradley, and Miss L. Day, towards the first three fasciculi ; for the future, specimens of any unusual form or supposed hybrid will be welcome on approval, on the chance of their proving suitable for incorporation, if a sufiicient quantity (about thirty-five sheets) from the same stock can be furnished." The first fascicle (1912-13) contains Ob THE JOURNAL OF BOTANY specimens of S. alba x fragilis, S. caprea, L. f. vel hybrida, *S'. caprea x myrsinites, S.o-urita, L., S.cinerea, L. (two examples), S. aurita X cinerea ; in the second (1913) are S. triandra, L., S. triandra, L. f. near subsp. Hojfmanniana (Sm.), S. aurita x purpurea, S. cinerea x purpurea, S. aurita x caprea, S. cajyrea x rcpens, S. cinerea, L. forma, S. Andersoniana x cinerea, S. Ander- soniana x j^^^ylicifolia. Offers of co-operation should be addressed to the Rev. Edward P. Linton, Edmondsham Rectory, Cranborne. The parts of Herr Carl Christensen's Index Filicum were noticed in this Jom-nal as they appeared during the years 1905-7. There is no need now to point out the remarkable value of this useful work ; for all students of ferns recognize its high merits, and the International Congress of Botanists at Brussels in 1910 selected it as the starting-point for the nomenclature of ferns. Herr Christensen has recently published a Supplement {Index Filicum. S^ipplementum, 1906-1912. Hafniae 1913 apud H. Hagerup. Pp. 133), divided into two parts — the first containing the names of 33 genera and subgenera and of 2611 species proposed during the years mentioned ; the second consisting of corrections of, and addi- tional synonyms to, several species adopted in the Index. — A. G. The recent parts of the Transactioyis of the Linnean Society include Mrs. A. Weber van Bossa's account of the Marine Algae (BhodophycecB) collected by Mr. J. Stanley Gardiner during the ' Sea-lark ' expedition to the Indian Ocean in 1905 (October, 1913) ; Professor Harvey-Gibson's observations of the morphology and anatomy of Mystropetalon (December, 1913) ; and an account of the cuticles of some recent and fossil Cycadean fronds by Mr. H. H. Thomas and Miss Nellie Bancroft, also published last December. Dr. Weber van Bosse's paper, which is illustrated by three plates, contains the description of two new genera, Oligo- cladus and Amphishetema, and of numerous new species ; Tapei- nodasya Ethelce commemorates Mrs. Gepp, who has unfortunately been prevented by illness from working out the collection. Some of the sections of Die Silsstvasser-Flora Deutsclilands, Osterreichs und der Schweiz, issued under the general editorship of Prof. A. Pascher, have recently been noticed in this Journal. Another section (Heft 14, Bryophyta. Jena : Gustav Fischer, 1914. Pp. iv and 222 ; figs. Price, Mk. 5-60, in paper cover ; Mk. 6-20, bound), the joint production of three experts, is now published, and comprises the following groups : Sphagnales (C. Warnstorf), Bryales (W. Monkemeyer), HepaticaB(V. Schiffner). The number of species comprised in the three groups is forty- eight, one hundred and forty, and sixty respectively. It is at first sight rather surprising to find so many mosses included in a list of water-plants. The explanation is that many species, which normally thrive in a dry habitat, have a capacity for adapting themselves to an aquatic life. The three authors have discharged their task conscientiously, and give careful descriptions of every genus, species, variety, and form, with keys to facilitate the work of identification. The figures are numerous, and many of them are original. — A. G, Journ.Bob. Tab. 530. p. HigLley del.et lith. West.NewmaTi imp . A. Mi-Lschleria angolensis S. Moore. B. Rhamphogyne rhynchocarpa S.Moore. .89 ALABASTEA DIVEESA.— Part XXIII. By Spencer le M. Moore, B.Sc, F.L.S. (Plate 530.) I. Vernoniace^ Africans nov^." Ethulia Scheffleri, sp. nov. Fmticosa, ramulis erectis crebro foliosis sessilibus vel subsessilibus lineari-oblanceolatis obtusis vel obtuse acutis basi coartatis margine serrulatis firme membranaceis cito glabris, capitulis subspbaeroideis pluriflosculosis ad apicem ramulorum corymboso-paniculatis, inflorescentiis polycepbalis foliis majoribus subsequilongis puberulis bracteis subulatis quam pedunculi proprii certe brevioribus prseditis, capitulis subspbae- roideis pluriflosculosis, involucri 4-serialis phyllis exterioribus parvulis ovato-oblongis acutis vel obtusis interioribus oblongis appendice scariosa purpurea dorso glandulis sessilibus lucentibus induta coronatis, coroliis exsertis, achseniis anguste cylindrico-tur- binatis truncatis 5-angulatis inter angulos glandulosis. Hab. Uganda, Lamuru station, 3000 m. above sea-level; Scheffler, 288. Frutex fide cl. detectoris 2 m. alt. Folia 3-8 cm. long., 8-13 mm. lat., accedunt pauca summa 2 cm. x 3 mm. nisi etiam minus. Inflorescentige +8x4 cm. Bracteae plurimae 2-4 mm. long. Pedunculi proprii capitulorum profecto evolutorum + 8 mm. long. Capitula pansa 7x7 mm. Involucri phylla extima 2 mm. inter- media 3-5 mm. intima 4-5 mm. long. Corollse tubus basi tenuis, superne anguste campanulatus, 2-5 mm. long., basi -3 mm. faucibus 1 mm. lat. ; lobi lineari-oblongi, tubo aegre aequilongi. Styli rami exserti, 1-5 mm. long. Achaenia vix 1-5 mm. long., brunnea. Foliage like that of some forms of E. conyzoides, but capitula larger with involucres unlike. Muschleria, Vernoniacearum gen. nov. (Plate 530.) Capitula homogama, tubuliflora, flosculis paucis. Involucrum cylindricum, phyllis pauciseriatis, arete imbricatis, angustis. Eeceptaculum leviter foveolatum, nudum. CoroUae regulares, tubo anguste infundibulari, limbo 5-lobo. Antherae basi sagittatae, auriculis obtusis anth. contiguarum connatis. Styli rami teretes, hirtelli. Achaenia 10-costata, compressiuscula angulataque, apice truncata. Pappus parvulus, cupuliformis, ore undulato. Caules erecti, bene foliosi, singuli vel plures rhizomati persistenti insi- dentes. Folia alterna. Capitula angusta, in glomerulos densos ad apicem ipsius caulis vel ramulorum brevium aggregata. Muschleria angolensis, sp. unica. Caule lignoso subsimplici apicem versus solummodo ramoso subtereti in longitudinem costato scabrido, foliis sessilibus linearibus apice mucronatis margine revolutis coriaceis supra scabridis subtus albo-tomentosis, glome- * The types of species described are in the National Herbarium. Journal of Botany. — Vol. 52. [April, 1914.] h 90 THE JOURNAL OF BOTANY rulis subsphseroideis vel corymbosis foliis perpaucis caulinis simi- libus nisi minoribus nonnunquam stipatis, capitulis 6-flosculosis, involucri 4-serialis phyllis lineari-lanceolatis longe acuminatis dorse carinatis interioribus quam exteriora longioribus omnibus rigidis puberulis, corollis exsertis, achisniis subturbinatis inter costas microscopice glandulosis. Hab. Angola, Forte Dom Affonso, Munongue and Kaconda; Gos&iveihr, 2907, 3092, 4144, 4325. Planta 1^-31 spithamea. Caulis 2-4 mm. diam. Folia solem- niter 1-5-4-5 cm. long., 1-5-3 mm. lat., pag. sup. olivacea. Glome- ruli subspbaeroidei, 2'5-3-5 cm. diam., corymbosi plerumque 6x5 cm. vel paullulum ultra. Capitula pansa usque ad 16 x 3 mm. saepe vero minora. Involucri phylla ext. 4-5-8 mm. long., int. 9-11 mm. Corollas purpureas ; tubus 7 mm. lobi 3 mm. long. Antheraj 3 mm. long. Stylus exsertus, superne incrassatus hirtellusque ; rami 2 mm. long. Achaenia 3 mm. long. Pappus •2 mm. alt. In appearance tbis is somewhat like a Corymbiuvi, the species of which have 1-flowered capitula, a different involucre and pappus with a lacerated edge. The generic name has been adopted in recognition of the work, especially on GompositcB, of Dr. Eeno Muschler. Vernonia (§ Lepidella) fontinalis, sp. no v. Herbacea, bi- spithamea, caule ex collo satis valido sparsim breviterque lanoso erecto simplici vel pauciramoso sericeo-tomentoso dein pubescente, foliis inferioribus rosulatis sessilibus oblongo-oblanceolatis obtusis chartaceis supra scabriusculis subtus appresse puberulis superi- oribus sparsis ceteris similibus sed minoribus, capitulis submedio- cribus pedunculatis subsessihbusve pluriflosculosis corymbos oligo- cephalos efiicientibus, involucri late campanulati sericeo-tomentosi 5-serialis phyllis oblongo-lanceolatis interioribus gradatim longi- oribus oblongis omnibus obtuse acutis, corollis subexsertis, achaeniis turbinatis prominenter 5-angulatis angulis setulosis faciebus glan- dulis lucentibus inspersis, pappi squamis anguste lineari-lanceo- latis acuminatis setis paucis scabriusculis stramineis corollas aequantibus. Hab. Angola, Kubango, in meadows at the source of the Kuartiri ; Gossweiler, 4180. CoUum usque ad 1 cm. diam., fibras simplices validas copiose emittens. Folia glandulis immersis creberrime induta, inferiora summum 5 cm. long., 13 mm. lat., superiora + 3 cm. x 5 mm., summa imminuta nee perpauca in involucri phylla non transeuntia, omnia in sicco viridia. Corymbi summum 10 x 5 cm., saepe vero breviores. Capitula pansa 11 X 15 mm. Involucri phylla char- tacea, extima 4 mm., intermedia 6 mm., intima 8 mm. long. Corollae ex schedis cl. detectoris rubro-violaceae ; tubus cylindricus, 6*5 mm. long. ; lobi lineari-oblongi, obtusi, apice setuliferi, 2-5 mm. long. Achaenia 2-5 mm. long, vel paullulum ultra, brunnea. Pappi squamae achaeniis circa aequilongae, setae 9 mm. long. This belongs to the group of species ranged round V. amhigua Kotsch. & Peyr. and V. Petersii O. & H., and of these most nearly VERNONIACRiE AFRICAN.^; NOVJ-] 91 approaching V. amhigua, from which the present plant differs chiefly in its rosulate, only puberulous leaves, elongated inflores- cences, larger heads, and acute not lengthily acuminate involucral leaves. Vernonia (§ Hololepis) Duemmeri, sp. nov. Caule erecto sursum ramuloso puberulo jam glabrescente, foliis subsessilibus ovato-oblongis apice mucronatis basi obtusis margine crenulatis membranaceo-coriaceis supra scabridis subtus brunneo-pubescenti- bus, capitulis mediocribus circa 60-flosculosis ad apicem ramu- lorum (nonnunquam perbrevium) solitariis foliis ultimis perpaucis stipatis, involucri subhemisphaerici pluriseriati phyllis lineari- lanceolatis (intimis lineari-oblongis) spinuloso-acuminatis interi- oribus gradatim elongatis tela araneosa subsparsa onustis, corollis breviter exsertis, achicniis cylindrico-turbinatis 10-costatis inter costas pilosis, pappo biseriali setis albis exterioribus quam interi- ores multo brevioribus. Hab. Uganda, Zinga and Ripon Falls ; Di'unmer, 35. [Also Grass land, Uganda ; Bev. C. Wilson, 72 in Herb. Kew.] Suffrutex vel frutex, ex schedis cl. detectoris 2-4 ped. alt. Folia pleraque 4-5 x l"5-2 cm., siccitate brunnescentia ; petioli lati, 2-3 mm. long. Capitula 14 x 18 mm., quidque foliis 1-5 nunc 2 cm. (vel etiam majus) long., nunc (sed raro) usque ad 5 mm. diminutis stipata. Involucri phylla extima 2 mm., intermedia 4-6 mm., intima 7"5-8 mm. long. ; horum acumen fuscum, leviter induratum, glabrum. Corollae tubus 7 mm. long., anguste infundi- bularis, triente inf. -3 mm. lat., faucibus 1 mm. ; lobi lineares, fere 3 mm. long. Styli rami 3 mm. long. Achaenia longit. 3 mm. paullulum excedens, optime costata, grisea. Pappi setse ext. fere 1 mm. long., int. 5 mm. Easily distinguished by foliage and involucres from the few species of this section. Probably conspecific with this is another Kew plant collected by A. Whyte in a journey from Nandi to Mumias. Vernonia (§ Xipholepis) paludigena, sp. nov. Rhizomate crasso sparsim fibroso caulem simplicem erectum supra basin nudam foliosum scabriusculum fulciente, foliis late oblanceolatis vel oblanceolato-obovatis apice mucronatis basi in petiolum brevem longe attenuatis margine subargute calloso-dentatis coriaceis nitidis utrinque scabridis costaque media pag. inf. puberula, capitulis submajusculis multiflosculosis in corymbos oligocephalos digestis, involucri subhemisphaerici circa 7-serialis phyllis lineari-lanceolatis longe acuminatis margine serrulato-ciliolatis chartaceis superne membranaceis viridibusque exterioribus saepe plus minus patenti- bus recurvisve, achaeniis anguste cylindricis 10-costatis basi callosis glabris, pappi squamis anguste linearibus acutis setis 2-serialibus scabriusculis dilute stramineis. Hab. Belgian Congo, in a marsh at Kabinda; Kassner, 2832. Herba fere metralis. Caulis 3-4 mm. diam., eximie striatus. Folia inferiora baud rosulata etsi perpauca infima approximata, 12- fere 20 cm. long., ultra medium 4 cm. lat., in sicco supra II 2 92 THE JOURNAL OF BOTANY viridi-bi'unnea subtus pallidiora; petioli circa 1'5 cm. long., supra canaliculati ; folia juniora gradatim diminuta. Corymbus circiter 12 X 8 cm. Pedunculi 3-5-12 cm. long. Bracteae paucae, lanceo- lataB, + 2 cm. long., ultimas capitula appropinquantes filiformes, 4 mm. long. Involucri phylla extima 5 mm. long., intermedia 10--14 mm., intima 18 mm., omnia dilute viridi-grisea. Corollas maturae baud suppetunt. Achaenia basi callosa, brunnea, 6 mm. long. Pappi squamas 2 mm. long., setas 9 mm. The affinity of this is with V. Melleri Oliver & Hiern, from which it differs in foliage and capitula. V. Melleri was redescribed by 0. Hoffmann under the name of V. scabrifolia, and referred to his § Laclmorhiza. To me this section seems unnecessar5^ for since the involucres and pappus yield such excellent sectional characters in this genus, I think it a pity to isolate a few species upon the fact of their having some woolly hairs upon their root- stocks: I agree, however, that if such a character is helpful in the grouping of the species of a large homogeneous genus, like Senecio, for instance, it may well be brought into use. Vernonia (§ Decaneuron) chlorolepis, sp. nov. Khizomate crasso sparsim piloso, caule erecto superne ramoso folioso etsi basi nudo valido tereti hispidule scabrido tandem glabrescente, foliis sessilibus oblongo-lanceolatis raro oblongo-obovatis acutis basi angustatis decurrentibus integris vel distanter denticulatis coriaceo-membranaceis pag. utravis scabris vel scaberrimis, capi- tulis magnis late campanulatis multiflosculosis ad apicem ramo- rum solitariis corymbum referentibus, involucri subhemisphaerici pluriserialis phyllis lineari-lanceolatis (intimis lineari-oblongis) mucronatis interioribus gradatim longioribus rigidis scabriusculis deorsum dilute stramineis sursum viridibus, coroUis exsertis, achgeniis cylindricis 10-costatis subtiliter pubescentibus, pappi setis 2-seriatis exterioribus abbreviatis scabridis rubiginoso-stramineis. Hab. Angola, Kubango, in thickets at Forte Princeza Amelia, and Kaconda at Landringo ; Gossweiler, 2331, 4253. Caulis circiter ^-metralis, 5 mm. diam., longitrorsum striatus. Folia solemniter 7-9 x 1-5-2-5 cm., summa diminuta et in invo- lucri phylla transeuntia, pallide viridia, utrinque prominenter nervosa ; reticulum laxum. Capitula matura 2'5-3"5 x 4-5 cm. Involucri phylla extima 8-10 mm., intermedia cii'ca 15 mm., intima usque ad 20 mm. long. Corollas coeruleae ; tubus extus sparsim glandulosus, angustus, sursum levissime ampliatus, 15 mm. long. ; lobi anguste lineari-oblongi, acuti, 8 mm. long. Antheras exsertae, 7 mm. long. Styli rami 4 mm. long. Achaenia 4*5 mm. long. ; pappi setas ext. circa 3 mm., int. 14 mm. long. V. carnea Hiern, to which this is allied, has somewhat different toothed leaves, much smaller heads and involucral leaves, although similar in shape, consistence, and colour, very much smaller. No. 2331 cited above has smaller heads (2x4 cm.) than the other, and its involucral leaves a little narrower. Vernonia (§ Decaneuron) ornata, sp. nov. Caule e rhizo- mate crasso fibrillifero ascendente robusto fistuloso in longitudinem eximie costato glabro, foliis maxima pro parte juxta solum approxi- VERNONIACEa: AFRICANS NOV35 93 matis sessilibus lineari- vel oblongo-oblanceolatis obtusis subdis- tanter calloso-denticulatis chartaceis utrobique scabriusculis paucis junioribus ceteris similibus nisi minoribus gradatim diminutis, capitulis magnis multiflosculosis corymbum oligocephalum pauci- bracteatum baud elongatum referentibus, involucri subbemi- sphterici 5-serialis phyllis lanceolatis (interioribus gradatim longi- oribus lanceolato-oblongis) acuminatis coriaceis superne vero saepius membranaceis fuscisque dorso scabriusculis margins scabriusculo-ciliolatis, corollis exsertis, achaeniis cylindricis basi callosis 10-costatis appresse setulosis, pappi setis 2-serialibus (paucis extimis abbreviatis) scabriusculis rubicundulo-stramineis. Hab. Angola, moist meadows in marshes of the Kiuriri near Kassuango and of the Cambambe at Kuebo ; Gossxoeiler, 3313, 3885. Circiter bispithamea. Caulis diam. adusque 5 mm. Folia in sicco viridi-grisea, summum 29 x 4 cm., ssepe 14-18 x 2 cm., caulina seniora + 10 x 1"5 cm., summa + 3*5 cm. long. ; costa media subtus optime eminens ; costae laterales planae ; reticulum laxum sat difficile aspectabile. Corymbus 2-5-cephalus, ± 10 cm. long. Capitula pansa 3'5 x 4-5-5 cm. Involucri phylla extima 5-10 mm. long., intermedia 15-20 mm., intima vix 30 mm. long. Corollas violaceae vel coeruleae ; tubus elongatus, anguste sub- cylindricus (superne leviter dilatatus), 25 mm. long. ; lobi anguste lineari-oblongi, obtusi, longit. 10 mm. paullulum excedens. An- therae exsertae, 6 mm. long. Styli rami 7*5 mm. long. Achaenia 7 mm. long. Pappi setae paucae 5-10 mm. long., complures 20-22 mm. The affinity is with V. carnea Hiern, which, however, is quite unlike it in several respects. Vernonia (§ Decaneuron) concinna, sp. nov. Herbacea, verisimiliter sat elata, caule erecto sesquimetrali vel paullo minus alto valido folioso glabro, foliis caulinis sparsis sessilibus lanceo- latis vel oblanceolato-oblongis obtusis margine distanter vel sub- distanter calloso-denticulatis basi breviter decurrentibus coriaceis glabris, capitulis mediocribus circa 25-flosculosis in paniculam terminalem angustam crebro bracteatam puberulam cito fere glabram digestis, involucri campanulati 7-serialis phyllis oblongis acutis interioribus gradatim longioribus intimis optime elongatis omnibus (apice fusco exemptis) tomento sordide albo vel dilute brunneo obductis, corollis breviter exsertis, achaeniis cylindricis 10-costatis inter costas glandulis lucentibus praeditis, pappi setis inter sese fere aequalibus (perpaucis interdum paullo brevioribus) plerisque apice curvatis ciliatis fulvis. Hab. Angola, Amboim-Novo Eedondo, near the Forte Quissaca ; Gossiveiler, 4480. Caulis 5 mm. diam., in longitudinem eleganter costatus, fuscus. Folia 9-10 cm. long., 2-3 cm. lat. basin decurrentem versus angustata necnon petioliformia, ima basi levissime dilatata, in sicco brunnea vel brunneo-viridia. Panicula circa 30 cm. alt., summum 7 cm. lat. ; axis cauli similis nisi angustior ; rami 94 THE JOURNAL OF BOTANY ascenclentes, distanter bracteati, apice oligocephali (raro monoce- phali), + 4 cm. long. Bracteae ex axi ortse pleraeque 1-5-2 cm. long., 5 mm. lat., margine denticulatge ; bracteae ramulorum plergeque 5-8 mm. long., summae imminutse in involucri phylla transeuntes. Capitula pansa 1"5 cm. long., totidemlat. Involucri phylla extima 2-5-3 mm., intermedia 4-6 mm., intima 10 mm. long. Corollae tubus anguste infundibularis, 10 mm. long., inferne vix 1 mm. lat., faucibus 2 mm.; lobi oblongi, 4 mm. long. Achaenia 3"5 mm., pappus 9 mm. long. Near V. Quartiniana A. Eich., but, among other features, the foliage and capitula of the two are very unlike. Vernonia (§ Stengelia) lafukensis, sp. nov. Erecta, veri- similiter suffruticosa, circiter bispithamea, caule sat valido simplici vel apicem versus rariramoso usque ad inflorescentiam folioso pubescente, foliis subsessilibus oblongo-oblanceolatis basi apiceque obtusis margine denticulatis pergamaceis utrobique in nervis puberulis glandulis immersis praeditis pag. sup. nervis planis vel subplanis pag. inf. (ut reticulum arctum) optime eminentibus, capitulis submediocribus breviter pedunculatis 12-flosculosis in corymbum brevem polycephalum bracteatum griseo-tomentellum subcongestis, involucri turbinati circa 7-serialis phyllis ovatis (intimis oblongo-obovatis) chartaceis perpaucis extimis appendice membranacea colorata lanceolata acuminata onustis ceteris appen- dice late ovata phyll. exteriorum acuminata phyll. interiorum saepe obtusa vel obtusissima et (costa media dorso excurrente) apiculata, corollis exsertis, achaeniis cylindricis pubescentibus obscure costatis, pappi setis pluribus (exterioribus paucis brevioribus) aliquanto complanatis apice saepe leviter clavellatis scabridis stramineis. Hab. Belgian Congo, Lafuka Eiver and in open fields at Lufongo ; Kassner, 2846 a, 2863 a. Folia inferiora 6-9 x l"4-2-2 cm., superiora gradatim diminuta ± 25 X 7 mm., in sicco griseo-viridia ; petioli summum 2-3 mm. long., validi, ut caulis fusci. Corymbi 3-6 x 5-10 cm. Bracteae foliis similes sed minores(+15 x 3 mm.). PeduQCuli proprii saepius circa 5 mm. long. Capitula maturata (flosculis inclusis) 18 X 8 mm. Involucri phylla margine dorsoque scepe pube grisea induta, pauca extima circa 10 mm. long., intermedia circa 12 mm. (inclusa appendice 6x6 mm.), intima 12-5 x 4 mm., horum appendix circa 2-5 x 2-5 mm. Corollae tubus inferne tubulosus, superne subito dilatatus, pars dilatata fere medium usque in lobos triangulari-obloDgos obtusiusculos divisa, pars tubulosa 8 mm. pars dilatata (inclusis lobis 2-5 mm. long.) 5'5 mm. long. Antherte partim exsertae, 5-5 mm. long. Styli rami 4 mm. long. Achaenia 2 mm., pappi setge pleraeque 7 mm. long. Allied to V. cardiolepis 0. Hoffm. and V. Britteniana Hiern, but differing from both in foliage and in the short crowded inflorescences. The sparsely hairy involucres with acuminate appendages are also good points by which it can be distinguished from the latter, while the involucral leaves of F. cardiolepis, besides having a rotundate appendage, are much narrower. VERNONIACE^ APRICAN^i: NOV^ 95 Vernonia (§ Stengelia) vallicola, sp. nov. Suffrutex caule ^-f-metrali erecto robusto basi nudo alibi crebro folioso densiuscule griseo-pubescente demum glabrescente, foliis sessilibus (summis subsessilibus) ovato-oblongis basi apiceque obtusis margine den- tato-seiTulatis subcoriaceis supra glabris subtus sparsim pubes- centibus costis (reticulo arcto hand exempto) pag. inf. maxime eminentibus, capitulis mediocribus circa 20-flosculosis in corym- bum subpaniculatum raribracteatum circiter 10-cephalum foliis duplo longiorem dispositis, pedunculis pedunculisque propriis griseo-pubescentibus his quam capitula plerumque longioribus, involucri cylindrico-turbinati 6-serialis phyllis exfcimis abbreviatis ovatis intermediis longioribus ovato-oblongis intimis oblongis vel late oblongo-linearibus omnibus chartaceis stramineis (exterioribus stramineo-brunneis) et lamina petaloidea obovata obtusa onustis, corollis breviter exsertis, achaeniis cylindricis basi callosis dense appresse setulosis, pappi straminei setis pluribus apice aliquanto dilatatis scabridis paucis exterioribus ceteris brevioribus. Hab. Angola, Kuito, in the valley of Fiengo ; Gosstveiler, 3781. Folia solemniter 5-6 x 2-2-5 cm., secus caulem subapproxi- mata (intermedia circa 15 mm. long.), pauca ima perpaucaque summa minora, sc. circiter 3-3'5 cm. long., hgec in bracteas transeuntia, in sicco griseo-viridia. Corymbi 9-10 x 8-10 cm. ; bracteae + 1*5 cm. long., paucae capitula appropinqantes anguste lineares, + 6 mm. long. ; pedunculi proprii plerique 1-5-3 cm. long. Capitula pansa 2-8 x 1-1-2 cm. Involucri phylla extima 4-5 mm. long., intermedia (appendice sibi ipsi aquilonga inclusa) 16 mm., intima 16 mm. long., horum appendix circa 3 mm. long. Corollae tubus attenuatus, sub apice subito in limbum campanu- latum fere usque medium lobatum ampliatus, longit. 2 cm. paullulum excedens. Antherae inclusa?, 6 mm. long. Styli rami 6 mm. long. Achgenia 3-4 mm. long., pappi setae pleraeque 12-13 mm. Easily distinguished from V. armerioides 0. Hoffm. by the leafy stems, differently shaped leaves with close and prominent reticulation and cylindrical-turbinate involucres. Vernonia (§ Stengelia) castellana, sp. nov. Herbacea, bi- spithamea vel paullo ultra, caule erecto siraplici basi nuda exempta sparsim folioso pube brevi densiuscula gaudiente, foliis subdistan- tibus (internodiis plerisque 2-3-5 cm. long.) oblongo-obovatis obtusis obtusissimisve basi in petiolum longum cuneatim coartatis margine undulatis vel denticulatis papyraceis supra scabriusculis subtus in nervis reticuloque eminente pubescentibus, capitulis inter minores circa 15-liosculosis in corymbum subscaposum longi- pedunculatum abbreviatum satis densum ordiuatis, bracteis per- paucis foliis multo minoribus, pedunculis propriis capitulis saepius aequilongis brevioribusve ut inflorescentiae rami tomentosis, invo- lucri campanulati pubescentis 4-serialis phyllis intus gradatim longioribus oblongo-ovatis (intimis oblongis) chartaceis appendice ovata (vel phyll. interiorum lanceolata) obtusa vel acuta onustis, corollis exsertis, achaeniis cylindricis obscure costatis setulosis, 96 THE JOURNAL OF BOTANY pappi setis 2-serialibus stramineis breviter barbellatis exterioribus abbreviatis angusteque squamatis. Hab. Angola, on the Cului Eiver, in pasturage at Forte Dom Affonso ; Gossiveiler, 2883. Foliorum limbus basi saepe obliqua, 13-15 cm. long. 4-5-6 cm. lat., in sicco griseo-viridis subtus paullo pallidior ; petioli validi, ima basi dilatati, obscure decurrentes, supra plani dorso rotundati sparsim pubescentes, + 4 cm. long. ; folia perpauca ultima + 5 x 2 cm. Pedunculus 30-40 cm. alt., pubescens. Bractese perpaucas vetustiores 2-3 cm. long., 3-4 mm. lat.; juniores capitula appropin- quantes lineares, + 5 mm. long. Capitula (corollis admissis) 12 X 6 mm. Involucrum 7 mm. long. ; phylla extima 2-3 mm. intermedia 6-7 mm. intima 9 mm. long. Corollae coBruleo-violaceae tubus superne amplificatus, 8 mm. long, (pars amplificata 2 mm.) lobi lineares, 3 mm. long. Antheras exsertae, 2-5 mm. long. Styli longe exserti, rami 2 mm. long. Achaenia 2"5 mm. long. ; pappi setae ext. 2 mm., int. 7 mm. long. In habit this is somewhat like V. viilancjiana S. Moore, in capitula V. lasiopus 0. Hoffm., near which latter species it should find a place. It is also close to V. prcemorsa Muschl., which, besides smaller and narrower leaves, has larger heads with broader involucral leaves. Vernonia (§ Stengelia) Anandrioides, sp. nov. Caule valde abbreviato filioso valido griseo-tomentoso rhizomate satis crasso tubera fusiformia hac atque iliac emittente suffulto, foliis rosulatis sessilibus late obovatis apice rotundatis basin versus angustatis et petiolum late alatum mentientibus margine repandis crenulatisve pergamaceis supra glabratis subtus in nervis griseo- tomentosis alibi pubescentibus glabrescentibus, capitulis medio- cribus pluriflosculosis paucis ex axillis foliorum solitatim oriundis, pedunculis scaposis erectis quam folia longioribus vel iisdem subaequilongis nudis vel bracteis perpaucis linearibus onustis dilute fulvo-tomentosis, involucri subhemisphaerici 4-serialis pubescentis phylhs paucis extimis bracteis similibus lineari-lanceolatis sursum attenuatis ceteris oblongo-ovatis (intimis oblongis) chartaceis appendice brevi lanceolata vel ovata colorata coronatis, corollis exsertis, achaeniis cylindricis basi callosis 8-costatis pube grisea obtectis, pappi setis paucis inaequilongis scabriusculis apice aliquantulum dilatatis breviterque barbellatis stramineis. Hab. Angola, Kubango, in gravelly pasturage near Forte Princeza Amelia ; Gossiveiler, 2132. Planta summum spithamea, saepe vero solummodo circa 6 cm. alt., habitu Plantaginis alicujus. Caulis 1-2 cm. alt. Folia viva solo applicata, 3*5-7 cm. long., ultra medium 3-4 cm. lat., supra in sicco brunneo-subtus griseo-viridia. Pedunculi 2-5-20 cm. long., apicem versus leviter incrassati ; bractese dum adsint + 8 mm. long. Capitula pansa 1-3-1-5 x 2 cm. Involucri phylla extima + 7 mm. long. ; intermedia (inclusa appendice 4 mm. long.) 10 x 4 mm. ; intima 11x3 mm., horum appendix circa 3 mm. long. Corolla vivide coeruleo-violacea tubus elon- gatus, attenuatus, 9 mm. long., superne subito campanulatim VERNONIACE^ AFRICAN.^ NOV^ 97 dilatatus; lobi oblongi, partis campanulatae dimidium adasquantes, 2-5 mm. long. Antheroe 4 mm. long., superne exsertoe. Stigmatis rami 3-5 mm. long. Achuenia 3-5-4 mm. long. ; pappi setae perpaucas exteriores 3-4 mm., ceterae summum 10 mm. long. Near V.pumila Kotsch. & Peyr., but with dissimilar coetaneous leaves as well as different involucres. The allied species V. nan- densis S. Moore has coetaneous but oblanceolate leaves, larger heads and involucral leaves, &c. Vernonia (§ Stengelia) campicola, sp. nov. Caule erecto crebro folioso angulato striatoque pubescente citissime puberulo, foliis amplis sessilibus oblongo-obovatis obtusis dimidio inf. gradatim attenuatis margine late dentatis membranaceis utrinque scabriusculis, capitulis majusculis multiflosculosis paucis (exempl. unici nobis obvii 4) ad apicem caulis corymbosis pedunculisque validis pubescentibus suffultis, involucri subsphaeroidei phyllis pluriseriatis interioribus gradatim majoribus chartaceis exterioribus ovatis interioribus late oblongis omnibus appendice ovata obtusa scariosa reticulato-nervosa onustis, corollis exsertis, achseniis cylindricis basi callosis 8-costatis pubescentibus, pappi setis aliquanto complanatis superne leviter dilatatis scabriusculis rubiginoso-stramineis paucis exterioribus ceteris brevioribus. Hab. Belgian Congo, Lufongo on open plains ; Kassner, 2845 a. Folia 18-23 x 3-5-6 cm. ; exstant vero minora, in sicco brunneo-virescentia, subtus pallidiora ibique sub lente glandulis lucentibus pilisque brevibus sparsim inspersa ; radicalia desunt. Corymbus 5x9 cm. Capitula 3-5 x 4 cm. Involucri phylla extima circa 10 mm., intermedia 15-20 mm., intima 25 mm. long. Corollse tubus elongatus, angustus, apicem versus subcampanulatus, 27 mm. long., hujus pars amplificata 4-5 x 2-5 mm.; lobi oblongo- lanceolati, 4 mm. long. Styli rami vix 5 mm. long. Achaenia fere 2 mm. long. Pappi setae ext. + 6 mm., ceterae + 10 mm. long. This belongs to the group of species clustering round V. in- signis 0. & H. and V. Calvoana 0. & H. The foUage and achenes are its chief peculiarities. Vernonia (§ Stengelia) Yatesii, sp. nov. Verisimiliter fruti- culosa ramis subtetragonis puberulis ramulis tenuibus pube cinerea densa obtectis mox puberulis, foliis parvulis oblanceolato-oblongis obtusis basi in petiolum brevem gradatim attenuatis margine apicem versus dentato-serratis firme membranaceis utrinque cito glabris vel fere glabris, capituHs pro rata parvis circa 25-flosculosis ad apicem ramulorum solitariis breviter pedunculatis pedunculis dense cinereo-pubescentibus, involucri campanulati phyllis circa 6-seriahbus interioribus gradatim paullo longioribus oblongis humectatis crassiusculis appendice brevi herbacea oblongo-ovata acuta onustis araneoso-pubescentibus, corolla3 tubo inferne attenuate superne dilatata, antheris basi acutis, achaeniis oblongis 4-5-costatis glabris, pappi setis comparate brevibus paucis (circa 18) 2-seriatis inferne subpaleaceis superne attenuatis valde inaequalibus scabridis sordide albis. Hab. Nupe, Nigeria; C. C. Yates. 98 THE JOURNAL OF BOTANY Folia l'5-2 cm. long., 4-8 mm. lat., in sicco olivaceo-nigra ; petioli 3-5 mm. long. Pedunculi adusque 4 mm. long. Involucra 7x8 mm. ; phylla intermedia 6 mm. long., 1-2 mm. lat. ; horum appendix + 2 mm. long., tenuia, difficile separabilia. Corollas puniceae, in toto 4 mm. long. ; tubus in feme -3 mm. superne fere 1 mm. lat. ; lobi oblongi, obtusi, 1 mm. long. Styli rami 2 mm. long. Acboenia 4 mm. long., in sicco dilute straminea, punctulis brunneis creberrime inspersa. Pappi setae ssepissime 1-5-3 mm. long. A curious little plant, with somewhat the look of a Centra- therum, but without the caducous pappus of that genus. The involucral leaves are so closely appressed as to make it impossible to separate them in the dry state, and not easy when moistened. Relatively to the achenes the pappus is short for Vernoiiia, and its hairs are few and in the dry state stand close together; moreover, when moistened they diverge slowly until they are at right angles to the achenes, no doubt a means whereby the achene is anchored when it falls upon moist ground. Under these circumstances I have doubts whether this should be placed in Vernonia, but I place it there provisionally, as there seems no more suitable genus to receive it. (To be continued.) MYCETOZOA FROM AROSA, SWITZERLAND. By Gulielma Lister, F.L.S. Arosa is in Canton Graubunden, about eight miles west of Davos. The village with its numerous hotels and sanatoria is finely situated at an altitude of 6000 feet, on the hillside high above the Plessur river, which flows north by a winding course to join the Rhine near Chur. The lower slopes of the steep valley are clothed with spruce woods ; above, the trees give place to moist meadows, alpine pastures, and mountain heights where snow often remains throughout the summer. The weather during the five weeks I spent at Arosa last summer was changeable. Three times heavy snow fell and covered all the land for a few days, when it was succeeded by rain and mist with intervals of brilliant sunshine. While searching for Mycetozoa in the latter part of June, I found but scanty and weathered remains of such species as Tricliia varia, Cribraria macrocarpa and Lycogala in the spruce woods. Later, when I had been joined by Miss Hibbert-Ware and Miss Schinz, the weather was more genial and we learnt the more favourable places in which to hunt. During the last week of July Plasmodium was found emerging from stumps and prostrate trees in many parts of the woods. On the alpine pastures, clumps of Cirsium sjnnosissimum, Senecio alpinus and Aconitum Nai)elUis afforded good hunting grounds. Amongst the fresh young foliage were many of the MYCETOZOA FROM AEOSA, SWITZERLAND 99 previous year's flower-stalks, decayed and hollow within, which, when split open, usually showed the tracks of plasmodium that had chosen this sheltered situation in which to creep, feed, and often also to form sporangia. Here Perichcena vermicularis abounded. Higher on the hills, where patches of winter snow still lingered or had recently lain, on the sodden brown turf were found such species as Physarum vernuvi, Didymium Wilczekii, Lcpidochrma Garestianum, and other cold-loving Mycetozoa. The following is a list of the forty-eight species we found, one of which appears not to have been previously described. Ceratiomyxa mucida (Mueller) Macbr. Abundant during the last weeks of July on spruce logs. The typical form only was found. Badhamia alpina, n. sp. Plasmodium pale yellow or yellow- ish white. Sporangia sessile, clustered or scattered, subglobose, or hemispherical on a broad membranous base, 0-5 to 0-9 mm. diam., grey, or dark brown and iridescent when without lime, usually seated on a dark horny layer of hypothallus over which are scattered deposits of lime-granules ; sporangium-wall membranous, fragile, with or without scanty deposits of lime granules. Capilli- tium a dense network of slender tubes, expanded at the nodes, and containing scanty deposits of lime. Spores greyish-lilac, very faintly warted, 10 to 12 jx diam., average 11 /u. Found in some abundance on and inside hollow scapes of Cirsium spinosissimum and Senecio alpinus, above Arosa, 7000 ft. alt. This species closely resembles B.foUicola Lister, from which it is distinguished by the pale colour of the plasmodium, the more hemispherical sporangia, and the pale nearly smooth spores. From B. panicea Eost. it differs in the yellow not creamy white colour of the plasmodium, the much closer network of the often limeless capillitium, and paler spores ; it differs also in the absence of any red colour at the base of the sporangium and of the strands of brownish-red hypo- thallus which usually characterize B. panicea. We had received previously three specimens of B. alpina from Graubunden gathered by Prof. A. Volkart, viz. one from Trimmis, at 5300 ft. alt., on her- baceous stalks, June, 1903, and two others from the Furstenalp at 5300 ft. and 6000 ft. alt. respectively, June, 1901, on dead raspberry twigs and old stalks of Veratnim album. M. Ch. Meylan has also gathered this species sevei'al times near Ste. Croix in the Jura Mts. A specimen found by Dr. E. E. Fries "in regione subalpina," 2600 ft. alt., at Areskutan, Jamtland, Sweden, in August, 1905, appears also to be B. alpina. All these specimens we formerly regarded as forms of B. foliicola with unusually pale and smooth spores before we realized how constant this character is, and that it is associated with pale plasmodium. Badhamia foliicola Lister. This first attracted our attention by numerous patches of bright orange plasmodium found creeping over twigs, fir-needles, grass and herbage under a spruce tree. By marking the spot we were able in a few days to collect fine clusters of the mature inconspicuous grey sporangia. The spores are violet-brown, 10 /t diam., and distinctly spinulose. This 100 THE JOURNAL OF BOTANY seems to be the first true record of the species for Switzerland, for in the light of our present knowledge the previous Swiss records should be referred to B. aliyina. Physarum viride (Bull.) Pars. One gathering only, on a spruce cone in the woods. P. NUTANS Pers. A single specimen found in the hollow of a spruce stump. P. FULVUM (Macbr.) Lister. A cluster of about fifty sporangia was found on a pile of spruce boughs in a steep wood. Except for a gathering made by M. Charles Meylan on Le Chasseron, Jura Mts, last May, this alpine species has not been found out of N. America, where it appears to be rare. M. Meylan's specimen, part of which he kindly sent me, has more globose and sessile sporangia than the type from Colorado (figured in Mycetozoa, pi. 66) ; they vary from yellowish -buff to white, or are yellow below and white above ; the lime-knots of the elastic capillitium are yellow and often unite to form a pseudo-columella. In the Arosa specimen the sporangia are obovoid on long or short yellow membranous stalks ; some are yellowish-chestnut, others are bright yellow all over, or yellow below and white above where the wall breaks up into fragile lobes ; the capillitium consists of an elastic network of yellow threads with numerous orange or orange-red angular lime-knots, containing irregular nodules of lime; the spores are rich brownish-purple, spinulose, 11-12 /x diam. This specimen shows a striking resemblance to Leocarjms fragilis Rost. both in the shape of the sporangia and in the capillitium and spores ; but although the colour of the sporangia varies in both these species, the walls of P.fulviim are membranous and rugose with included deposits of lime-granules, and show nothing of the polished cartilaginous layers characteristic of Leocarpus fragilis. P. VERNUM Somm. On Cirsiiim scapes. FuLiGO SEPTiCA (L.) Gmelin. One weathered sethalium found on a spruce stump. DiDERMA NivEUM (Rost.) Macbr. Found in abundance, but mostly in a weathered condition, on turf near melting snow. D. Trevelyani (Grev.) Fries. Two clusters found on Cirsium scapes on the alps. The sporangia show no trace of columellae. D. radiatum (L.) Lister. A curious development consisting of a dozen brownish-buff sporangia was found on a chip of spruce wood. They are sessile or on very short flesh-coloured stalks, or form short curved plasmodiocarps. The lime-granules are dis- solved from the cartilaginous sporangium-\valls which are marked with the cavities where the granules lay ; the columella is hemispherical, the capillitium colourless, and the spores purple- brown, 8 to 11 /x diam. I have since received specimens of typical stalked D. radiatum gathered at Arosa in August by the Hon. Terence Bourke. DiDYMiUM DiFFORME (Pers.) Duby. Both large and small MYCETOZOA PROM AROSA, SWITZERLAND 101 plasmodiocarps were found in many places on the alps on dead Cirsium stalks. D. WiLCZEKii Meylan. This alpine species was found abundantly on the alps, and showed great diversity of form. The largest gathering was on dead scapes of Aconitum Napellus at an elevation of about 8000 ft. near the Altein Furka Pass ; it consisted of some hundreds of sporangia clothed with flat discoid scales formed of closely cohering crystals of lime. A remarkable variety was obtained on the Ochsenalp. The specimen consists of two depressed plasmodiocarps, about 13 mm. long ; the superficial crystals form a continuous crust, and are easily brushed off from the glossy purple-brown sporangium-wall ; a columella is represented by the thickened spongy base of the plasmodiocarp ; the spores are unusually large, 13-16 jj. diam., purple-brown, closely and minutely spinulose. The most striking feature is the capillitium, the threads of which are pale purple, sparingly branched, about 2 /x diam., marked with a few dark bead-like thickenings, and also with three to four slender close-set spiral bands ; these either wind smoothly round the thread, or together form a prominent spiral : the direction of the spiral is similar to that on the elaters of Trichia. We have been accustomed hitherto to regard the development of spiral bands on the capillitium as a character of generic importance, found only in the Calonemeae division of the Mycetozoa. It is, therefore, rather disconcerting to have perfectly regular spiral bands formed on all the capillitium threads of this Didymium which in every other respect resembles D. Wilczekii. Lepidoderma Carestianum Eost. A few thick-walled plasmo- diocarps were found on turf near melting snow, and on Girsium scapes. The capillitium of one specimen shows an unusual struc- ture ; the threads are pale yellow, simple below, branched above, and consist of a central strand surrounded by a sheath broken up into long or short segments ; such a structure usually occurs in the capillitium of CoUoderma, but I have not seen it in any other species. Colloderma oculatum (Lippert) G. Lister. On decayed spruce stumps, on both sides of the Plessur valley. Fragments of mossy wood with sporangia were brought back to England and kept moist under a bell-jar for four months, during which time fresh sporangia continued to appear at one point or another over the surface. Besides a growth of moss the wood was coated with a gelatinous green alga, with which Colloderma is usually associ- ated. Through the alga the plasmodium emerges as scattered sulphur-yellow beads, which soon secrete a gelatinous envelope ; in many cases the mature sporangium pushes out through this envelope, enclosed only by a delicate iridescent membrane which readily breaks and allows the spores to be scattered. Stemonitis fusca Eoth. On spruce logs. S. ferruginea Ehrenb. On spruce logs, found only in a weathered condition. 102 THE JOURNAL OF ROTANY S. HERBATiCA Peck. On a spruce log, weathered. CoMATRiCHA NIGRA (Pers.) Schrad. Fairly abundant on fallen spruce boughs. C. TYPHOiDEs (Bull.) Eost. var. heterospora Eex. On a spruce stump. Lamproderma violaceum (Fr.) Eost. var. Caresti-e. Eather abundant on and inside hollow Cirsmm scapes. The sporangia are all shortly stalked, some are globose, others are ovoid; in one group of sporangia the sporangium-walls may either persist in large flakes, or break up into fragments adhering to the tips of the capillitium : the capillitium forms a dense intricate net- work, and is purplish-black except at the extreme tips, which are colourless ; the spores are very minutely spinulose and either very dark grey, slightly pale on one side, 11-5 to 14 /x diam. ; or, in one gathering, they are pale grey, 10 to 12 [i. These specimens differ, M. Meylan tells me, from his L. atrosporum by the capillitium being too dense ; in L. atrosporum the spores can be freed from the sporangia at a breath. Enerthenema papillatum (Pers.) Eost. On decaying stumps. Cribraria aurantiaca Schrad. On rotting spruce logs. C. MACROCARPA Schrad. Abundant. On one tall stump the sporangia extended over an area of several square feet, but were old and mouldy. C. purpurea Schrad. Found on the same stump as the largo growth of G. macrocarpa. The sporangia measure about 3 mm. in total height , they are old, and have turned a dull crimson. A similar change from the typical reddish-purple colour to crimson has taken place in a few sporangia of a large gathering kindly sent me from the Jura Mountains by M. Meylan. C. RUBiGiNOSA Fries. A fine but weathered development was found on a spruce stump. The sporangia measure from 3 to 3'5 mm. in total height : the deep cups are marked below with numerous, close-set oblique or curved lines of thickening, and above with a close reticulation ; the sporangium-walls seem to be persistent throughout. In Mycetozoa, ed. 2, p. 177, the specimen from Berne, quoted under this species, should be referred to M. Meylan's new species C. femiginea.--' This differs from C. rubiginosa in the more piriform shape of the sporangia, in the shallow cups being marked wath about twenty strong ribs, and in the rusty-red spores. The colour of the spores is the chief character that distinguishes C.ferruginca from C. macrocarpa. DiCTYDiuM CANCELLATUM (Batsch) Macbr. var. alpinum. Found in several places on stumps. The largest growth covered an area of three square inches on the under side of a spruce log : the sporangia are erect or inclined on stout stalks 2 mm. high : they resemble typical var. alpinum in having only twenty to thirty ribs to the sporangial net, but differ in the colour of the ribs being * See Annuaire du Conservatoire et dn Jardin hotaniqiie de Genh'e, 1913, p. 319. MYCETOZOA FROM AROSA, SWITZERLAND 103 deep red instead of brown ; in some sporangia the ribs branch and anastomose to form a Cribraria-\ike net, in others the ribs send off obhque branches on either side, as in Schrader's ilkistration of Dictydium venosuvi.''- LiCEA FLEXUOSA Pors. Found developing from rosy Plasmo- dium and also in a mature condition on spruce stumps and boughs. L. MINIMA Fries. Abundant on a few much decayed spruce stumps. The sulphur-yellow plasmodium continues to emerge and form sporangia for some days over the same area of wood, so that one may find all stages of growth intermixed, varying in colour from yellow to bright chestnut or brownish-black. The dark granules dotted over the inner surface of the margins of the lobes of the sporangium walls are not free " plasmodic granules " such as we see in Gribraria, Dicttjdiwn and Lindhladia, but prominent, often irregular warts or outgrowths from the walls themselves. L. pusiLLA Schrad. Fairly abundant on decayed spruce stumps. The colour of the emerging plasmodium is watery drab. I am not aware that this has been recorded before. When the sporangia are still young and pale the outlines of the areolae and lobes into which their walls will ultimately divide are clearly defined as a net of dark lines. This species can be distinguished from L. minima in the field by the spores being black instead of a reddish-brown colour in mass. TuBiFERA FERRUGiNOSA Gmelin. A cluster of sporangia was found emerging from a spruce stump in pinkish-yellow plasmo- dium, which matured indoors. Lycogala epidendrum (L.) Fries. On spruce stumps. Trichia pavoginea (Batsch) Pers. Abundant on spruce logs and stumps. The elaters in some sporangia have the spiral bands replaced partially by ring-shaped thickenings. T. scABRA Rost. Two large developments were found on spruce logs. T. persimilis Karst. On spruce stumps. T. VARiA Pers. Abundant on wood. T. coNTORTA (Ditm.) Rost. var. alpina. On Cirsiwn scapes. T. DECiPiBNS (Pers.) Macbr. On spruce stumps. T. BoTRYTis Pers. On spruce logs and stumps. Hemitrichia Karstenii Rost. A group of brown sporangia was found on a heap of spruce boughs. H. abietina (Wig.) Lister. On decaying spruce. The bands on the capillitium are two or three, arranged in a loose spiral, and often interrupted by ring-shaped thickenings. H. leiotricha Lister. A single shortly-stalked sporangium was found on a chip of spruce wood. The structure of the sporan- gium-wall with its comma- or ring-shaped thickenings and deposits of olivaceous refuse matter and the colour and markings of the spores are typical. The capillitium is remarkable in having the * See Nova Genera Plcmtarum, pi. iii. fig. 6. 104 THE JOURNAL OF BOTANY spiral bands on the threads running in the reverse direction from that in Trichia, i. e., they pass from the right below to the left above when the thread is seen horizontally. This is the first instance that I have seen of reversed spiral bands in any species of the Mycetozoa ; the nearest approach is in Oligonema flavidmn Peck, where the rows of minute warts that are studded over the capillitium are often arranged to form rows passing round the threads in the reversed spiral direction. Of the cause or possible significance of such a reversal we know nothing. Its exceptional occurrence draws our attention to the constancy with which the usual type of spiral occurs. Aecyria denudata (L.) Sheldon. A large development of crimson sporangia was found on a decaying spruce log. A. INCARNATA Pors. Developed indoors from white Plas- modium found on a spruce stump. A. NUTANS (Bull.) Pers. Found emerging as a mass of creamy-white plasmodium from a spruce stump, and matured indoors. It was probably slightly injured when it was cut off the stump, for, although the spores are normal, the capillitium is irregular, varying from 5 to 8 /x diam., and studded all over with stout conical spines. Perich^na vermicularis (Schwein.) Eost. Abundant on and inside hollow Cirsium scapes on the alps. The slender buff sporangia form simple or net-like plasmodiocarps ; the plasmo- dium, instead of being watery-white, as we had previously seen it, is rosy-red; the veins of bright red sclerotium were conspicuous when the stalks were split open and examined. Margarita metallica (Berk. & Br.) Lister. Found twice on spruce wood ; the sporangia were either solitary or clustered. Dianema corticatum Lister. Fairly abundant on decaying roots and stumps of spruce, both in the rosy plasmodium stage and as mature sporangia. Prototrichia metallica (Berk.) Massee. Two sporangia found on decaying spruce roots ; both are brilliantly iridescent and mounted on brownish-yellow stalks about 1 mm. high ; the colour of the spores in mass is olive-brown, instead of the usual pink or brownish-pink of freshly formed spores. After we left Arosa, the Hon. Terence Bourke and Miss Jasmine Bourke kindly sent me collections of Mycetozoa made there by them during the month of August. Of the fifteen species which they found, the following five had not been seen by us at Arosa, viz. : — Leocarpus fragilis (Dicks.) Eost. DiDYMiuM MELANOSPERMUM (Pcrs.) Macbr. CoMATRiCHA TYPHOiDES (BuU.) Eost. The typical form. Cribraria piriformis Schrad Several developments of per- fectly developed sporangia, with spores varying in colour from bright reddish-brown to purplish-brown when seen in mass. Arcyria FERRUGiNEA Sautcr. 105 TWO HEPATICS NEW TO BRITAIN. By W. E. Nicholson. During the autumn of 1912 I found in several stubble fields in the neighbourhood of Lewes a small species of Biccia which generally had some traces of violet colouring about it and was most frequently ciliate. It bore some general resemblance to E. Warnstorfii Limpr., but it seemed constantly different in the more compact rosettes with shorter wider branches. This seemed to point to R. commutata Jack, and on my submitting the plant to Dr. Schiffner, he confirmed it as that species, remarking that it was rather more compact than the plant from the original locality and that from Dalmatia, but that this might well be accounted for by the habitat, as morphologically and in the spores it agreed well with these. B. commutata is no doubt closely allied to B. Warnstorfii, and the differences may to some small extent be accounted for by differences in the habitat of the two species, B. Warnstorfii being perhaps more frequently found in lighter soil than B. commutata ; but, on the other hand, the differences are retained on the cultivation of both species in the same soil. In Sussex the two species sometimes grow in the same field, and Mr. H. H. Knight has found the same thing in Gloucestershire. On the Continent B. Warnstorfii is credited with a northern range, being principally recorded from Northern Germany, while B. commutata has a distinctly southern dis- tribution. Both plants will probably be found to have a much wider distribution in Britain than is at present suspected. In the autumn of 1912 B. commutata was much more abundant in Sussex than B. glauca, with which it might perhaps be confused in the younger stages, though if transverse sections be cut of mature fronds of the two species there is no possibility of mistaking one for the other. The following diagnosis, largely borrowed from that of Dr. K. Miiller, may be of use in dis- tinguishing the plant (K. M. Mus. Hep. 1 Abt. p. 191). Monoicous. Thallus small, dark vivid green, rarely reddish, flat, two or three times forked, 2-7 mm. long and 1-1-5 mm. broad. Branches linear, oval or ovate with margins here and there stained with red, usually inserted almost rectangularly, truncate to emarginate at the ends, with or without marginal cilia and with a narrow channel only at the ends of the branches; further back the upper side of the thallus is convex. Frond section one and a half to three times as broad as thick, ellipsoid in the older parts with rounded margins wider towards the apex, slightly convex below, flat above with a short obtuse sinus and bow-shaped towards the sides. Cells of the epidermis thin- walled, spherical, without mamilla. Ventral scales colourless or reddish violet, soon disappearing. Ostioles rising slightly above the upper surface of the thallus. Spores brown 80-85 fx with a distinct yellow and notched margin, closely papillose, areolte consequently indistinct, 8 fj. wide, 6-8 visible in the diameter of the spore. Journal of Botany. — Vol. 52. [April, 1914.] i 106 THE JOURNAL OF BOTANY Habitat. East Sussex over a wide area ; East Kent {E. M. Holmes), West Gloucestershire and Worcestershire {H. II. Knight). Dr. Miiller describes a var. acrotricha Lev. with marginal cilia, but I have generally found a few cilia on the Sussex plant when freshly gathered. These frequently disappear when the plant is cultivated under moist conditions, but they reappear when the plant has finished its growth or is cultivated under drier and more exposed conditions. I am inclined therefore to regard the presence of cilia in this species as a normal feature of the plant, representing a state in the life of the individual frond rather than a true variety. The Sussex plant is rarely of a vivid green except when growing under very moist conditions ; most frequently it is of a rather pale glaucous colour, recalling that of B. glauca, though it usually has some trace of violet both in the frond itself and the ventral scales. It is probable that some of the plants which I referred to the var. siihinermis of B. glauca in the past may have belonged to the present species. The other plant, also rather a critical form, is a species of Fossomhronia, which I gathered in Babbacombe Bay, near Torquay, in March, 1913. I noticed in the field that the plant had a rather stout stem, strongly convex on the under side with hyaline or, occasionally, brownish rhizoids, unlike the violet rhizoids of all the described British species, and the microscope revealed the fact that the papillae on the spores, which were otherwise like those of F. ccBspitifonnis, showed an occasional tendency to anastomosis, so as sometimes to make the surface of the spore very irregularly areolate. These characters seemed to indicate F. Husnoti Corb., but on my submitting the plant to Mr. S. M. Macvicar he pointed out that the spores were considerably larger than those of typical F. Husnoti, 43-53 /x, as against 40-45 fx in that species, and that their surface was less distinctly areolate. These differences I was well able to observe by comparing my plant with a specimen of F. Husnoti from Florence, which I owed to the kindness of Dr. Schiifner. On my submitting the Babbacombe plant to Dr. Schiffner, he pronounced it to be certainly F. Husnoti, and he apparently attached considerable importance to the frequent presence of three spiral elaters. In view, however, of the departures from the type above noted I have ventured to describe it as a new variety : — FossoMBRONiA HusNOTi Corb. var. anglica. A F. Husnoto cliffert sporis grandioribus •043-053 mm.,papillis minus rcgulariter anastomosantibus foveolas paucas valde irregulares formantibus. Hab. Moist banks by the sea, Babbacombe Bay (TF. E. N.), near Llandovery, S. Wales (H. H. K). F. Husnoti and F. ccBspitiformis are very closely allied species, and it might seem as reasonable to describe the present plant as a variety of the latter as of the former, but in a genus where the violet rhizoids are such a marked feature of the other European species, it seemed worth while to draw attention to a form with hyaline or brownish rhizoids, a feature which I have found maintained on cultivation of considerable material. The stem, SHORT NOTES 107 moreover, is very strongly developed on its under side so as to make it rather more subterranean than in the other species, and no doubt the frequent presence of three spiral elaters is a character of some importance. I have, however, found this feature not infrequently present in Sussex forms of i^. ccBspitiformis, though it is less usual in forms from S. Europe. SHORT NOTES. Ptilota plumosa Ag. (p. 77). — For over forty years I have been the possessor of a fine and unmistakable specimen of Ptilota plumosa in my herbarium, which was sent me by the late Mr. Henry Goode, as collected by himself " near Falmouth, Cornwall, summer of 1871." I knew Mr. Goode only by correspondence, but was assured of his reliability and absolute sincerity, and as one who only distributed algae of his own collecting, invariably off the coasts of S. Devon and Cornwall. I believe he passed away in the early eighties, at a good old age. His abode was near Plymouth. I may add that I have never myself found P. plumosa Ag. in any Welsh station save that of Path-y-Pistill, Holyhead, where in the summer of 1883 I found many most perfect and beautiful fronds, usually floating. I have searched in vain for it at Penluaen Mawr, Llanfairfechan, Llandudno, Barmouth, &c. — J. Cosmo Melvill. Valerianella eriocarpa Desv. — Eeferring to my notes as to the occurrence of this plant in the Isle of Wight ( Journ. Bot. 1912, 231 ; 1913, 288), it may be worth recording that on this date (March 2nd) there are thousands of young plants in the same habitat, where in 1912 they were abundant, and where last year I could only find, after long searching, one or two plants. If the Botanical Exchange CIuId or any botanist would care to have specimens I would gladly dry some as soon as the plants are in fruit. — Frederic Stratton. Aberdeenshire Plants. — Mimulus mosckatus, reported (Journ, Bot. 1911, 370) as being found by me in Haughton Wood, Afford, Aberdeenshire, is increasing greatly since that date. — Linaria repens still grows at Auchindoir, Aberdeenshire (whence it is recorded in Dickie's Guide (1860) ), where there was once a cottage garden. The stems attain the length of three feet. Cattle had eaten the longest stalks overhanging the dyke, before I (in August last) took careful measure. — William Wilson. PoLYPORUs SQUAMOsus. — Somo vory early fruit bodies of Polyporus sqiuunosics have appeared recently in the " Backs " at Cambridge. My attention was called to one of these by Mr. Maltby ; this was a fructification produced inside the hollow trunk of an elm in St. John's College Backs. On further exami- nation of trees, other fructifications were found — several on a horse chestnut near the elm tree, and three or four more on an 108 THE JOURNAL OF BOTANY old elm stump (behind King's College) which was known to be badly infected with the fungus. The fructifications appeared in all probability during the second part of February, and were past maturity in early March. Generally they were rather small. The usual period for the fructification to appear is from May to September, as reported by Buller. I have, however, recorded fructifications in the Cambridge district as late as the middle of November and now as early as February. The winter has of course been mild in Cambridge. — S. Eeginald Price. REVIEWS. A Flora of Norfolk loith Papers on Climate, Soils, Physiography and Plant Distribution, by Members of the Norfolk and Norwich Naturalists' Society. Edited by W. A. Nicholson. Demy 8vo, cloth, 214 pp. 2 maps. Price 6s. Mr. Kirby Trimmer's Flora of Norfolk was published in 1866 ; it was never a satisfactory book and has been for many years out of print, and the Norfolk and Norwich Naturalists' Society has done well to prepare a new one. Mr. W. A. Nichol- son had for some years been collecting material for such a work, and the Society acquired his MSS. which, " with much additional information compiled from the work of botanists, resident and non-resident," is now issued in a neat volume. The book, so far as one who is practically unacquainted with the county is at liberty to express an opinion, is very carefully done. The small (though clear) type employed enables the volume to appear in a more pocketable form than is usual in similar works, and there is an absence of the irrelevant matter which sometimes finds its way into local floras ; space has been saved, too, by a free but intelligent use of abbreviations. The scattered records in this Journal and elsewhere seem to have been carefully swept up. The four districts into which the county was divided by Mumford in his list in "White's Directory— da,te not stated — and adopted by the Society in 1869 is followed here ; a brief account is given of their characteristics. The other important matters indicated in the title are duly discussed at reasonable length ; there has evidently been a commendable desire not to be prolix or discursive. We regret, however, the absence of what has become a pro- minent— sometimes too prominent — feature in our local floras — some account of the history of the botany of the county and of those who have taken part in its investigation. Such an account adds not only to the value but to the interest of the book ; and Norfolk has been so especially favoured by botanists that the omission is the more to be regretted : J. E. Smith, W. J. Hooker, Dawson Turner, among illustrious names ; Pitchford and Crowe among those who grew and studied Willows, as was fashionable in botanical circles towards the end of the eighteenth century. A glance through the Biographical Index of Botanists will reveal A FLORA OF NORFOLK 109 many more, including some about whom information is badly wanted, which should have appeared in the roll-call of botanists which we expect to find in a local flora. We also miss the descrip- tive and critical notes which often lend more than a local value to works such as this, and which indicate observation of a higher type than that which a mere record conveys. We note the presence of a considerable number of names in local use, and the omission of the more absurd of the " book- names " which — e.g., in Carex — often disfigure books of this class. We are glad, too, that all the names, whether Latin or English, are placed in one index. Nor must we omit to mention what is evidently a very full and careful list of Mosses and Liver- worts prepared by Mr. W. H. Burrell, with the assistance of Mr. H. N. Dixon ; the Fungi are altogether omitted. There are two excellent maps. It will be seen from what has been said that we have in Mr. Nicholson's book a solid though not an exciting contribution to our knowledge of British local botany — a knowledge the gaps in which are being steadily filled up. Pflanzenphysiologie : Versuche unci Beohachtungen an hoheren und niederen Pjianzen einschlusslich Bacteriologie und Hydro- hiologiemitPlanktonkunde. By K. Kolkwitz. Jena: Gustav Fischer. Pp. 258. 12 coloured plates, 116 text figures. 9 Mark. The present work has grown from the author's Pflanzeri- physiologische Versuchen zu Ubimgen im Winter, published in 1899, and is a course of physiology arranged for the students of the Berlin University and the Agricultural Academy. The first part, consisting of sixty pages, deals with the nutrition of the higher plants — chlorophyll and its functions, turgidity and os- mosis, sugar and reserve food material, proteins, respiration, water and air. The experiments are well-devised, though most of them are familiar ; the exposition is clear and reference is made to recent literature. The principal part of the book deals with Cryptogams. After a description of the microscope and its acces- sories, the various phyla are considered in order, beginning with the Myxomycetes and ending with the Ferns. Much of the matter cannot be said to be physiological in the ordinarily accepted sense of the term. Methods of collection, culture, and examina- tion, as well as a description of the habitat, are given where possible. The physiological significance of various processes in the life-history is well described in many instances. The most interesting portion of the book is the section " Algen, Plankton und Okologie der Gewiisser," as was to be expected, seeing that the author is a recognized authority in hydrobiology. The short descriptions of many plants might have been left out with advan- tage. In some cases the cryptogams so treated are commonly met with in the neighbourhood of Berlin (and of London) and might therefore claim place, but Sargassum, Macrocystis, &c., seem out of place in such a book. This portion of the work would 110 THE JOURNAL OF BOTANY prove an excellent introduction to the study of cryptogamic botany. The book is unique in the stress which it lays upon the cryptogamic section of the plant world, a section known principally in many of our English " schools " through the intricacies of the vascular anatomy of Pteridophytes — present and past. It is well printed and well illustrated. ^ -p Icones Orchidearum Austro-Africanarum Extra-tropicarwn ; or, Figures with Descriptions of Extra-trojncal South African Orchids. By Harry Bolus, F.L.S., Hon. D.Sc. (Cape). Volume iii., demy 8vo, tt. 100, with text. Wesley. Price £1 10s. net. It is satisfactory to learn that the late Mr. Bolus's important work is to be carried on by Mr. H. M. L. Bolus, the Curator of the Bolus Herbarium. The present volume was apparently left in MS. by the author of the preceding instalments, as his name alone appears on the title-page, and the preface, while acknow- ledging the help of Mr. F. Bolus, who has drawn nine of the plates and made additions of various kinds to others, makes no reference to any collaborator in the literary portion. Of the plates in the volume before us, 36 have already appeared in the author's Orchids of the Cape Peninsula, which is now out of print. Among the genera represented, Disa has 36 plates, Satyriitm 12, Disperis and Pterygodinm 11 each ; the other genera included are Acrolophia, Eulophia, AngrcBCum, Bartholina, Huttoncea, Holothrix, Hahenaria, Brachycorythis, and Ccratandra. The descriptions correspond with those of the previous volume noticed in this Journal for 1912 (p. 28). There is a new species of Holothrix — H. Beckii, named after its discoverer and sole collector ; and a hybrid — Satyriuni coriifolium x carneum — described from three living specimens, found growing widely apart, but always among the parent species. Floral Evolution : with Particular Reference to the Sympetalous Dicotyledons. By H. F. Wernham, B.Sc. Neio Phytologist reprint, No. 5. Pp. viii + 151. Price 3s. Cambridge: at the Botany School. The above work is based upon a course of lectures delivered by Mr. Wernham at the Chelsea Polytechnic in the summer of 1910, and was published in its present form in the pages of the Neio Phytologist. The author states that his purpose has been to give a connected account of one of the larger groups (namely, the GamopetalcB or Metachlamydea or Sympetalce) of flowering plants, with continual reference to a definite evolutionary story, having for its motif a few broad biological principles applicable to the phylogeny of the Angiosperms generally. " In the endeavour to preserve the continuity of this evolutionary story, it has been necessary to hasard suggestions of affinities and lines of descent which may be by no means generally acceptable ; and the student is warned that in such cases the suggestions are purely tentative." FLORAL EVOLUTION 111 Mr. Wern ham's primary contention is that the GamopetalcB are polyphyletic in origin ; and on this matter the reviewer is in full sympathy with the author. Many botanists have put forward more or less detached suggestions to the same effect ; but none perhaps has stated the case more whole-heartedly. At the same time, the reviewer confesses that some of the particular sugges- tions of afUnities are regarded in the light of the author's warning of their purely tentative nature. The work is very readable and very well written, and Mr. Wernham has performed a useful service in drawing the attention of botanists to a department of their science which, in these days of callunetums and heterozygotes, is too apt to be altogether for- gotten. We are bound to confess, however, that the methods of study of floral morphology have not kept pace with the develop- ment of technique in other branches of botany. In early and mid-Victorian days researches on the morphology of the parts of the flower were not unfashionable ; but since the microtome came into vogue there has been a great falling off in work of this character. Yet it is undoubtedly the case that many of the problems discussed in Floral Evolution are incapable of solution without assistance from the microtome and all its concomitant paraphernalia. The reviewer looks forward to the time when problems of floral morphology will once more become a vogue, and when modern methods of research will be applied to this branch of botany, which is at once most interesting and most important from the evolutionary point of view. C. E. M. BOOK-NOTES, NEWS, dc. We are glad to see that the Times is taking up the question of the extermination of primroses near London, and indeed further afield; an article published in its columns for March 13 is followed by a letter from Mr. H. Kowland-Brown, of Harrow Weald, who writes : — " For something over forty years I have watched the annual invasion of the tramp dealer in our woods here, and along the once exquisite Middlesex lanes in which, also, the wild hyacinth flourished abundantly, and the nightingales sang in the high hawthorn hedges. Primroses, hyacinths, high hedges, and night- ingales are now all gone from the lanes, and every Sunday in the season motors bring down troops of indiscriminate ' collectors ' to harry the few remaining beauty spots in the woods left by the itinerant vendors, for whom at least there is some intelligible excuse. I have seen half-a-dozen cars drawn up by the roadside by a tiny copse known from my childhood as 'the Bluebell Place,' and immortalized as such on the canvas of English painters. A few years more, and — the wood being no longer preserved for game — the hyacinths will have followed the primroses; for the most distressing feature of the work of destruction is to be seen upon the road itself strewn with broken spikes of bloom, and bulbs dragged up by the greedy picker, to whom locked gates and thorn 112 THE JOURNAL OF BOTANY hedges offer no obstacle sufficient. It is just the same with the willow ' palms ' — flowering this year happily too soon, at least for the Easter holiday crowd — every bush on the common, every branch by the way broken down and disfigured as high as hand may reach." It is many years since we expressed surprise that the Selborne Society had not made its influence more felt in the matter of plant preservation : the formation of an influential "Plant Protection Section" is an important step in the right direction, but so far we are not aware that any definite results have followed : meanwhile the question becomes yearly of greater urgency. Mr. Brown concludes his letter by an appeal to " the local authority," which might, we would think, at least control or modify the efforts of the roadmen who throw upon the roadside turf the parings and hedge trimmings they annually remove, leaving them there to destroy what yet remains of grass ; and who scrape the hedge banks and hedge bottoms to the great benefit of the nettles, which profit by the destruction of other plants and usurp their place, to the great detriment of the charm of the countryside. The volume on Wild Floivers which is contributed by Mr. Macgregor Skene, B.Sc, to Messrs. Jacks' series of "People's Books " is arranged according to colour — a scheme which is doubtless intended to save trouble in identification, but which can hardly be considered scientific. Nor are we sure that it attains its object, for folk see colours very differently — we should not ourselves class Eed Campion as purple or Water-Plantain as rose ; nor is it easy to see why Bladderwort is placed among "flowers rarely found or very inconspicuous," seeing that it is described as " fairly common " and the flowers as " large yellow." Mr. Skene recommends Babington's Manual and Bentham and Hooker's Handbook, but omits Hooker's Students' Flora, which is more useful than either. As a companion to the Handbook he recommends " Smith's Illustrations of the British Flora "; by this is intended the volume of figures by Fitch, prepared for the illustrated edition of Bentham's earlier Handbook (which Bentham and Hooker's replaced), which were issued as a separate work, with supplementary figures by Mr. W. G. Smith, in 1880: from this are taken the numerous excellent illustrations in Mr. Skene's book, which form its most attractive feature. The descriptions are careful and accurate and the volume is a wonderful sixpenny- worth. The February number of the Journal of Genetics contains a " Preliminary Note on the Genetics of Fragaria," which includes experiments with F. vesca and notes on garden hybrids ; it is illustrated with one plate and cuts. Mr. E. S. Salmon writes " On the Appearance of Sterile 'Dwarfs' in Hnnmlus Lnp7ilns," with two plates. The other papers deal with zoological matters. Corrections. — In our last issue, p. 72, line 8 from bottom, "not" should be inserted before " hitherto " ; on p. 86, line 14 from bottom, the words "of Cambridge (not Dublin, Mr. Wilson !) " should read " of Dublin, formerly of Cambridge." Journ. Bot. Tab. 531. ><) r< Cm O CL, o c - .,• CD ^ _- -c L' « iJ a:' .2 M S ■CJ3 i- Q.:^ '•^ ^ rt ■" . = : § =* <^ -o ^ w ~= c .:; o •• o g o - ^'5 cv-r •~ •= w « '-' a: S < 5 w 2 m^ Ho < i < i^ 2 = SO O ': O.-. 113 A NEW FOSSIL COEEMA. By Clement Keid, F.K.S., & Eleanor M. Reid, B.Sc. (Plate 531.) The Empetrace(2 form so small an order, with only three genera and about six species, that the discovery of a new species fossil in Britain is worth recording. The fruits in question were found by us in 1904 in the pre-glacial deposits of Pakefield, in Suffolk. Their curious shape and mode of attachment made us then refer them to some unknown species of ViburniLvi, and as such they were described in 1908.''' The great difficulty in cutting sections of these pyritised fruits from the Cromer Forest-bed made us also less careful to examine their internal characters than we should have been. A few years after the discovery of the Pakefield specimens two endocarps belonging to the same plant f were found at Tegelen, in Dutch Limburg, in deposits somewhat more ancient than those of Pakefield, and probably of about the age of our Norwich Crag. These also were figured by us as belonging to this unknown species of Viburnum. The study of a still older Pliocene flora in Dutch Limburg, with many fruits and seeds belonging to unknown genera, has lately necessitated the systematic examination of the flora of the Pala3arctic Region, as represented by the fruits and seeds in the Kew Herbarium. In the course of this work we came across the small genus Gorema, and were at once struck by the resemblance of its endocarps to our unknown fossils. A close comparison of the recent fruits with our fossils left no doubt whatever that they belong to the same genus, though the fossils represent an extinct species. This Corema happens also to be the first extinct plant that has been recognized in the Cromer Forest-bed, though several other plants found in that deposit no longer live in Britain. Of the two living species of the genus, the one nearest our fossil, C. alba, is found on the coast of Portugal and Spain, and in the Azores. The other, C. Conradi, belongs to the coastal region of the New England States in North America. Thus not only does our fossil show a former wider distribution of the genus, but it is the first plant now specially belonging to the Atlantic province to be found fossil in the North Sea basin. As regards the conditions under which the extinct Corema grew, we can only say that the two deposits in which it has been found were both in all probability laid down as river alluvium within a few miles of salt water. This would agree with the habitat of the two living species, both of which are confined to coastal regions. * Litinean Sociehfs Journal, Botany, xxxviii. p. 215, figs, lb-11. t Verslagen Kon. Akad. Wet. Amsterdam, 1910, p. 267, figs. 30, 31. Journal op Botany.— Vol. 52. [May, 1914.] k 114 THE JOURNAL OF BOTANY Corema intermedia, sp. nov. Planta fossilis nisi fructus ignota. Drupa . . . endocarpiis 3, longit. 2-2-2-5, latit. 2-0-2-5 mm.; monospermis ovatis osseis extus incrassatis, dorso irregulariter costatis, facie canaliculatis. Testa tenuis membranacea alveolata, alveolis irregulariter transversis. Endocarps 3, ovate, hard, bony, thick-walled, irregularly ribbed longitudinally on the dorsal surface, channelled on the ventral face and facetted at an angle of about 120". Seed with a thin membranous testa with cells arranged in irregular zigzag trans- verse lines. Length of endocarps 2-2-3-0 mm. ; breadth 2-0- 2-5 mm. In size our endocarps are midway between those of C. alba and C. ConracU. In shape they resemble C. alba, though they are considerably smaller, more ovate, and less tumid. To C. Conradi, which has minute fruits, they bear little resemblance. The very peculiar cells of the membranous adherent testa are like those of C. alba, but the adhesion of this testa to the endocarp makes it difficult to examine in the fossil, and we have been unable to photograph its very characteristic structure, though this can be seen in places in the interior of the specimen shown in fig. m. NOTES ON BKITISH PLANTS. By C. E. Moss. II. Eanunculus obtusifloeus. It seems to be generally admitted by modern systematic botanists that Bammculus baudotii Godron (1839) " and B. con- fusus Godron (1848) should be reduced to a single species. They were, in fact, so reduced by Syme (1863), by Sir J. D. Hooker (1884), and by Rouy & Foucaud (1893). Syme united Godron's two plants under the name B. baudotii, and gave each a varietal name: B. baudotii Godron became B. baudotii var. vulgaris Syme (1863) ; and B. confusus Godron became B. baudotii var. confusus Syme (1863). Hooker named the same aggregate species (i. e., B. baudotii Godron ampl. Syme) B. marinus, and mentioned that he could not verify the characters of the varieties as constant. Rouy & Foucaud used the same aggregate name as Syme, and reduced B. confusus Godron to a subspecies. Mr. N. E. Brown (1891) also expressed the opinion that B. baudotii Godron and B. confusus Godron are only forms of one and the same species. This aggregate species {B. baudotii Godron ampl. Syme) has a west-European and Mediterranean distribution ; but it is not so markedly Atlantic in its range as the allied species B. homio- phyllus Tenore, 1830 {=B. lenormandi Schultz 1837), B. tripar- titus DC. (1808), andi?. hololmcus Lloyd (1844): this last is not known to be a British plant ; but one feels that it must occur in southern England, or at least in the Channel Islands. * Full citations are given later on in this note. RANUNCULUS OBTUSIFLORUS 115 Apparently western Europe is exceptionally rich in species of Batrachian Ranunculi. In choosing the name B. baudotii for his aggregate species, Syme passed over an earlier one, Batrachium obtusiflorum S. F. Gray (1821). It was, of course, customary in this country at that time to follow " the Kew rule " ; and consequently Gray's name would have been ignored by Syme even if he had seen it, since it was placed in a different genus. There was also the name B. tripartitus Nolte (1826) to consider ; but this name was pre-occupied by B. tripartitus DC. (1808) for another legitimate species. Gray's plant {B. obtusiflorum) was founded on B. tripartitus var. obtusiflorus DC. (1818), and de Candolle's variety, in its turn, on an illustration by Petiver {English Herbal, t. 39, fig. 1, 1713). This plant of de Candolle's was an addition to the B. tripartitus DC. (1808) which became B. tripartitus var. ynicranthus DC. (1818). It is clear that B. tripartitus var. obtusiflorus DC, being founded on one of Petiver's plants, is British ; and I am satisfied that Petiver's figure must be referred to B. baudotii Godron ampl. Syme. It is equally clear that Gray's plant is the same as de Candolle's var. obtusiflorus, and therefore that Gray's trivial name (being the earliest) must be utilised for Syme's aggregate species ; and this is the but of the present communication. Godron, when founding his B. baudotii, remarked that the plant showed a closer affinity with B. tripartitus DC. than with B. aquatilis L. emend. ; " and it certainly is the case that it is an interesting connecting link, as regards both its characters and its distribution, between the two species B. tripartitus DC. and B. aquatilis L. emend. On the whole, it seems justifiable to retain these as three distinct species, as is done by Hooker fil. and by Rouy & Foucaud. There seems to be a real (though a small) gap between each of them ; and if they are not kept as separate species, the resulting aggregate becomes so unwieldy that more confusion is caused by their union than by their separation. In particular, the interesting distribution of the segregate forms becomes lost sight of or at least obscured. As to the subdivisions of Syme's aggregate B. baudotii (or B. obtusiflorus, as it must now be named), Syme's varietal names appear to be the earliest ; and, if this is so, they must be adopted by those who regard Godron's two plants as being of varietal rank and who follow the international rules of botanical nomen- clature. Godron's two plants are undoubtedly very closely related ; and few, if any, students of Batrachian Ranunculi will desire either that they should be kept up as distinct species, or placed apart from each other as subdivisions of different species. Fries's plant {B. marimom) too is extremely close to those of Godron : in fact, it was reduced by Godron to a variety of his B. baiidotii ; and both Hooker fil. and Rouy & Foucaud follow * This is R. aquatilis L. Sp. PI. 556 (1753) excl. vars. ; = R. diversifolius Gilibert (1782) emend. Kouy Sz Foucaud (1893), non Schrank (1789). K 2 116 THE JOURNAL OF BOTANY Godron in this particular matter. Fries's plant indeed seems to be merely a submerged state of B. baudotii Godron destitute of floating leaves, as maintained by N. E. Brown (1891). The name " tripartihis " has been applied to at least five different plants (c/. Koch in litt. cited by Godron, 1839), namely, to the two British forms or varieties of B. tripartihis DC, to B. ohtusiflorus, to a form of B. aquatilis L. emend., and to a form of B. trichophyllus Chaix emend.''' Of the two British forms or varieties of B. tripartitus DC, one has been referred by Messrs. Groves (1907) to B. lutarius Bouvet (1872) = B. kitarium Eevel (1865). I have seen no authentic specimen of Revel's plant, which has actually been placed by French botanists under B. homiophyllus Tenore \—B. lenormandi Schultz). However, all the British plants in question which I have seen appear to be better placed under B. tripartitus DC than under B. Jiomiophylhis Tenore { = B. lenor- mandi Schultz) ; and this opinion coincides with the view of N. E. Brown (1891). One may perhaps be permitted to doubt if this plant of Revel has really been found in the British Isles ; or is it the case that Revel's plant should be placed under B. tri- partitus DC ? The latter is by no means an untenable view, if one may judge from Revel's description and figure. One may now allude to B.petiveri Koch. Koch first used this name in Sturm Deutschl. Fl. (1840), and again in the second edition of his Synopsis (1843). B. tripartitus Nolte (1826) and B. tripar- titus var. ohtusiflorus DC are cited by Koch as synonyms. N. E. Brown (1891) states that there is a specimen of B. tripar- titus Nolte in Herb. Mus. Brit., and that he agrees with Hiern (1871) in referring Nolte's plant to B. confusus Godron ; and this agrees with Hiern's allocation (with which I fully concur) of de Candolle's var. ohtusiflorus. I think too the figure in Sturm (82, 2) may also be referred to the same species, i. e., to B. ohtusi- florus. However, the B. petiveri Cosson and Germain Fl. Env. Paris 10, Atlas, t. 1, fig. 5-6 (1845) is not Koch's plant, being referred (and no doubt correctly) to his B. hololeucos by Lloyd in the various editions of his Fl. de I'Ouest. In Koch's Syn. ed. 2 (1843), B. petiveri is subdivided into two varieties, namely, var. minor and var. major. The var. minor is the B. petiveri Koch in Sturm (1840), discussed above. The var. major is regarded (erroneously, I think) by Hiern (1871) as the same as B. triphyllos Wallroth (1840). The latter plant seems rightly placed by Rouy & Foucaud (1893) under their B. diversi- folius { = B. aquatilis L. emend.). It is closely related to B. heterophylhts Babington (1855) non Wiggers (1780) nee Hooker fil. (1884), and is a rare plant. It occurs in the Channel Isles, though I have not seen the restricted plant of Wallroth from the British Isles proper. * Chaix (1786) established his E. trichophyllus on No. 1162 of Haller's Hist. Stirp. Helv. ii. 69 (1768) : from this it is necessary to exclude Haller's var, /3 which is R. fceniculatus Gilibert (1782) = i?. circinatus Sibthorp (1794). But see also F. N. Williams in Journ. Bot. xlvii. (1908). RANUNCULUS OBTUSIFLORUS 117 B. petiveri var. major Koch (1843) is referred by Kouy & Foucaud, on the other hand, to the particular form of the species B. trichojihylhis which was named B. radians by Eevel (1853). This appears to be a form of B. tricliophylUis with floating leaves, and is, in some ways, a link connecting B. obtusiflorus and B. trichophyllus. Koch, in the synonymy of his B. petiveri var. major, cites his own B. aqiiatilis var. tripartitus (1835) ; and, judging from the authentic illustration of this in Sturm Deutschl. Fl., Kouy & Foucaud would seem to be quite correct in their determination. The plant I regard as B. radians Eevel is locally abundant in the ditches of the fens of Cambridgeshire : it seems to be essentially identical with what passes for " B. godroni Grenier," but I have seen no authentic description of the latter plant. Grenier promised one in 1850, in Schultz Arch, de Fl. ; but I have been unable to find that the promise w^as ever fulfilled. The synonymy of B. obtusiflorus and its two west-European varieties is as follows : — E. OBTUSIFLORUS comb. nov. ; B. tripartitus var. obtusiflorus DC. Syst. Nat. i. 234 (1818) ; Batrachium obtusiflorum Gray Nat. Arr. Brit. PL ii. (1821); Hiern in Journ. Bot. ix. 69 (1871); B. tripartitus Nolte Novit. Fl. Holsat. 51 (1826) non DC; B. petiveri Koch in Sturm Deutschl. Fl. 82, 2 (1840) ; Batrachium marinum Fries Fl. Suec. Mant. iii. 51 (1842); B. petiveri var, minor Koch Syn. ed. 2, 13 (1843) ; B. baudotii [Godron am pi.] Syme Eng. Bot. i. 24 (1863) ; N. E. Brown in Eng. Bot. ed. 3, Suppl. 13 (1891); Eouy & Foucaud Fl. France, i. 65 (1893); Batrachium salsuginosum Dumortier in Bull. Soc. Eoy. Belg. ii. 217 (1863) ; B. marinus Hooker fil. Stud. Fl. ed. 3, 5 (1884). (a) E. OBTUSIFLORUS var. vulgaris comb. nov. ; B. baudotii Godron Essai in Mem. Soc. Eoy. Nancy, 21, fig. 4 (1839); Babington in Ann. Nat. Hist. ser. 2, xvi. 395 (1855)!; Batrachium baudotii Van den Bosche Prodr. Fl. Batav. 7 (1850) ; B. baudotii var. vulgaris Syme op. cit. 25 (1863) ; form baudotii Hiern ■'' op. cit. 69 (1871). Icones : — Babington in Eng. Bot. Suppl. t. 2966, as B. baudotii. Exsiccata : — Billot, 2802, as B. baudotii ; Fries, ix. 28, as B. marinum; F. Schultz (Herb. Norm.), 404, as B. baiidotii; Wirtgen, ix. 436, as B. baudotii. (b) E. obtusiflorus var. confusus comb. nov. ; B. confusus Godron in Grenier & Godron Fl. France, i. 22 (1848) ; Babington op. cit. 394 (1855) ! ; B. baudotii var. confusus Syme Eng. Bot. i. 25 (1863) ; form confusus Hiern op. cit. 69 (1871) ; B. baudotii subsp. confusus Eouy & Foucaud Fl. France, i. 66 (1893). Icones : — Fl. Dan. t. 1993, as B. tripartitus ; Koch in Sturm Deutschl. Fl. h. 82, t. 2. as B. petiveri ; Syme Eng. Bot. i. t. 23. * Mr. Hiern's "forms," or " ultimate forms," as he terms them on p. 44 (1871), have no definite systematic grade; and consequently it is incorrect to cite Mr. Hiern as the author of these "forms" when they are given definite rank, such as species, subspecies, race, variety, ox forma. 118 THE JOURNAL OF BOTANY Exsiccata : — Billot, 3801, as B. confusus ; Fries, xv. 28, as B. confusum ; Wirtgen, ix. 437, as B. 2)etiveri. In the above remarks it has, of course, been necessary to refer to plants which are related (either botanically or in nomen- clature) to B. ohtusiflorus ; and below will be found the species (as I understand them) of the section Batrachium of western Europe, with the full citations of the various plants referred to in the present communication. The citations are placed after the names of the species to which I consider they belong. A dash is placed before those names which, in my opinion, ought to be distinguished as subspecies or varieties. In recognising nine west-European species of the section Batrachium, I am following several leading modern authorities. Syme (1863) united B. aqua- tilis and B. tricJiophyllus; but otherwise his species were the same as those given below. Hooker (1884) and Eouy & Foucaud (1893) adopted precisely the species which are here given. Bentham (Handbook Brit. Fl. 59, 1858) reduced them all to one which he named " B. aquaticus Linn." [sic] ; but I know of no modern systematic botanist in Europe who would now accept Bentham's view of species ; whilst Babington (Man. ed. 9, by H. & J. Groves, 1904) divided the section into sixteen species. 1. E. HEDERACEUS L. Sp. PI. 556 (1753) ! 2. E. HOMioPHYLLus - Tonore Fl. Nap. iv. 338 (1830); B. ccenosus Gussone Suppl. Fl. Sic. Prodr. fasc. ii. 187 (1834) ! ; Godron in Grenier & Godron Fl. France, i. 19 (1848)!; B. lenor- mancli Schultz in Bot. Zeit. xx. 726 (1837) ! — ? Batrachium lutarium Eevel in Act. Soc. Linn. Bordeaux, xxv. 413, t. 4 (1865); B. Uitarius Bouvet in Bull. Soc. Angers for 1871, 96 (1872). 3. E. TRiPARTiTus DC. Icon. PI. Gall. Ear. 15, t. 49 (1808) ; N. E. Brown op. cit. 13 (1891) excl. syn. Knaf ; B. iripartitus var. micranthus DC. Syst. Nat. i. 234 (1818) ; Batrachium trijmrtitum Gray Nat. Arr. Brit. PL i. 721 (1821) ; form tripartitus Hiern op. cit. 68 (1871) ; — form intermedius Hiern op. cit. 67 (1871) excl. syn. Knaf ; B. lutarius ['? Bouvet, loc. cit.] H. & J. Groves in Journ. Bot. xlv. 452 (1907). 4. E. HOLOLEUCUS Lloyd Fl. Loir.-Inf. 3 (1844) ; B. petiveri Cosson & Germain Fl. Env. Paris, 10, Atlas, t. i. fig. 5-6 (1845) non Koch. 5. E. OBTUsiFLORUS Moss. SsB abovB. 6. E. AQUATiLis L. Sp. PI. 556 (1753) excl. vars. ; Godron Essai in Mem. Soc. Eoy. Nancy, 24 (1838) ; Koch, Syn. ed. 2, 12 (1843) ; B. heterojjhyllus Wiggers Fl. Hols. 42 (1780) ; Hooker fil. op. cit. 5 (1884) ; B. diversifolius Gilibert Fl. Lituan. iii. 262 (1782) ; Eouy & Foucaud Fl. France, i. 63 (1893) ; non Schrank; Batrachi^Lvi heterophyllum S. F. Gray op. cit. 721 (1821) ; — B. triphyllos Wallroth in Linnaea, xiv. 584 (1840) ; form tri- The justification of this citation may form the subject of a future note. MISCELLANEA BRYOLOGICA 119 phyllos Hiern op. cil. 69 (1871) excl. syn. Koch. ; — Batrachium heterophijllnm''- Fries Summ. Veg. Scand. 140 (1846) non S. F. Gray; B. heterophyllus Babington op. cit. 393 (1855)!, non Wiggers nee Hooker til. 7. E. FLUiTANs Lamarck Fl. France, 6d. 2, iii. 184 (1778). 8. E. TRicHOPHYLLUs Chaix in Villars Hist. PI. Dauph i. 335 (1786) emend. ; Hooker fil. op. cit. 6 (1884) ;— E. drouctii [Schultz ex] Godron in Grenier & Godron Fl. France, i. 24 (1848) ! ; — B. aquatilis var. tripartitus Koch Syn. ii. (1835) ; B. petiveri var. major Koch Syn. ed. 2 (1843) ; Batrachium godronii Grenier in Schultz Archiv. Fl. 172 (1850) nomen; Grenier Eev. Fl. Mont. Jura, 25 (n. d.) nomen ; B. radians Eevel in Act. Soc. Linn. Bordeaux xix. 120, fig. 1 (1853) ; form radians Hiern op. cit. 99 (1871) incl. form godronii. 9. E. ciRCiNATUs Sibthorp Fl. Oxon. 175 (1794) excl. syn. L.t In conclusion, a few words may be added regarding the terminations of some of the trivial names. In consulting the literature of these water-crowfoots, one has been struck by the great want of uniformity in two matters. One sees indifferently B. hololeucus and B. hololencos, B. tripliyllus and B. triphyllos. On the whole, one prefers the Latinised form in all such cases. Again, there is much inconsistency in the use of one "i " or two " i's " as genitive terminations. One sees indifferently B. baudoti and R. baudotii, B. godroni and B. godronii, B. hachi and B. bachii, B. lenormandi and B. lenormandii. The former would seem to be the more correct form ; but the recommendations of the botanical congresses tend to keep the matter in an unsettled state, for they illegitimately ask for two " i's " when a word ends in a consonant, and are inconsistently content with one " i " if the word ends in " -er." It seems, indeed, as if the recom- mendations with regard to the construction of botanical names need revision by a committee of scholars. MISCELLANEA BEYOLOGICA.— III. By H. N. Dixon, M.A., F.L.S. (Continued from Journ. Bot. 1913, p. 330.) Sematophyllum acutirameum (Mitt.), a "composite" species. In the course of working out a collection of mosses made by Eev. C. H. Binstead in Ceylon, it became necessary to study the above plant. It was described by Mitten (Muse. Ind. Or. p. 106) as Stereodon aciitirameus ; the localities given being " In Ceylon, Gardner \ In mont. Khasian., ad Moflong, in pinetis, Griffith ! " * Probably this ought to be cited as a subspecies ; but the point is con- fused by Fries placing his " Batrachium " heterophylluvi under ''Ranunculus " aquatilis. t Cf. Williams in Journ. Bot. xlvii. 15 (1908). 120 THE JOURNAL OF BOTANY On examining the material at Kew, I found a great confusion of specimens. Neither Gardner's plant nor Griffith's is there represented ; the specimens of Sevi. acutirameum determined by Mitten are Thwaites's C. M. 239 and 2396 (with others of Bed- dome's from the Nilgiris). It was at once evident that there were at least two species represented here, and it was necessary to ascertain which, if either, was Mitten's species. Mrs. Britton was kind enough to search Mitten's herbarium, and to send me specimens of Gardner's plants (Ceylon, No. 110 — the type, and Ceylon, No. 71) as well as of Thwaites's No. 239. Gardner's two plants are identical, and on examination prove to be the same thing as S. monoicum (Bry. Jav.) Jaeg. The authors of the Bry. Javanica, it may be recalled (ii. 208), say of their species, " affinis videtur Stereodon acutirameus Mitt." Authentic specimens of S. monoicum agree exactly with Gardner's plant, and it may be pointed out that there is absolutely nothing in either the figures or the description of Hijimum monoicum at variance with the diagnosis given by Mitten. Why then did the authors of the Bry. Jav. describe the Javan plant as new ? The answer is no doubt to be found in the note which accompanies Mitten's description of his S. acutirameiis, " S. Braunii simillimus, sed foliis angustioribus et florescentia diversa." Now, as >S. Braunii (C. M.) is not by any means closely like S. acutirameiim (i. e. S. monoicum), and has leaves very distinctly narrower than in Mitten's species (it is in fact one of the markedly narrow-leaved species), it was natural that the authors of the Bry. Jav. should consider their plant M'ith its widely oval leaves as distinct, in spite of the applicability to it of Mitten's diagnosis. The problem remains, what led Mitten to describe his type from Ceylon as having leaves narrower than in S. Braunii (C. M.) ? This is easily solved, since it is quite clear that Mitten entirely misunderstood S. Braunii, as is plain on referring to Gardner's No. 784, which Mitten records (M. Ind. Or., I.e.) as Ster. Braunii, but which is quite a different plant with distinctly wider leaves, and is, I have little doubt, /b'. Nietnerianum (C. M.) Jaeg. {Hyimum Nietnerianum C. M. in Linn. 1869-70, p. 64), with the description of which it entirely agrees. The seta is smooth above, not scabrous as in S. Braimii, the capsule very small. We may go a step further, and find the explanation, with great probability, of Mitten's misunderstanding of S. Braunii. It must be recollected that Mitten had not, at that time, the Bry. Jav. with its excellent figures to consult, and had to depend for his knowledge of S. Braunii on C. Muller's description in the Synopsis (ii. 687), and on any available specimen. Now the specimen of " Hy2J. Braunii MiilL, Herb. Dz. & Mb., Java," in Hooker's Herb., is probably the only specimen which would be available to Mitten at that time, and is no doubt that on which his conception of S. Braunii would be based. But this unhappily is not S. Braunii at all. The leaves are much wider and more shortly pointed than the figures in Bry. Jav., and the seta is smooth, or only indis- tinctly roughened at apex. It is a small specimen, and I do not MISCELLANEA BRYOLOGICA 121 venture to name itr' but it is at least very close to S. Nietnerianum (C. M.), of which Miiller writes : " Habitu H. sigmatodontii vel Braunii . . . differt . . . ab hoc pedunculo laBvi, foliis multo latioribus ovatis (nee anguste lanceolatis ubique convolutisj." It also resem- bles in appearance S. Gedeanuvi (C. M.). Now there is another specimen at Kew, under S. Gedeanum, which from the mounting and labelling was clearly sent out from Leyden at the same time as the specimen labelled H. Braunii in question. The labelling of the two is as follows : — H. Braunii MiilL, Herb. Dz. & Mb., Java. H. Gedeanum Miill., Herb. Dz. & Mb., Java. The writing on the two is almost in facsimile, and is no doubt in the same hand. Now the specimen labelled " i?. Braunii'' is, as I have said, a very difierent plant from the true Braunii, with more erect, much wider leaves. The other specimen, "jff. Gedeamim,'" consists of two tufts, one of which is certainly entirely S. Braunii, the other contains some stems also of S. Braunii, with another species which may well be S. Gedeanum. It appears extremely probable that the labels have been inter- changed in the process of putting up these duplicates ; in any case, the erroneous naming is with little doubt the cause of Mitten's misunderstanding of S. Braunii and the consequent train of errors. Upon them he based his conception of S. Braunii, and his record of Ster. Braunii (Gardner, 784) from Ceylon, which is really S. Nietnerianum ; and also described his Ster. acutirameus as having leaves narrower than in S. Braunii ; which later, no doubt, led the authors of the Bry. Jav. to consider their Javan plant as distinct, and to describe it as H. monoicum. What now are Thwaites's 239 and 239 &, determined by Mitten as Ster. acutirameus ? The specimen at Kew of 239 consists of two species, in four tufts, which I have marked a, b, c, d. Of these a, b, c are identical with Mitten's type (No. 110, Gardner), i. e., S. monoicum (Bry. Jav.). The plant marked d is different, and agrees with 2396 (which is the same plant in the Kew and British Museum specimens). This is a tall plant with narrow, longly convolute, spreading leaves, not infrequently secund ; it is dis- tinctly synoicous ; the seta, roughened above, is about 1 cm. long, the capsule very small (1 mm. long, including the peristome). The perichsetial bracts are erect, gradually pointed and slightly denti- culate. It is clearly, I think, S. sigmatodontiiim, which has not, I believe, been recorded from Ceylon ; but its existence there is quite in accordance with its recorded geographical distribution. It agrees in all its general characters, and the synoicous inflores- cence is, I think, quite conclusive. This, however, does not exhaust the whole question of the Ceylonese " S. acutirameum." Among the specimens sent by • The examination of herbarium specimens of Sematophyllum is rather difficult, except with a good deal of dissection, since it is always necessary to ascertain tlie nature of the inflorescence, while the seta and perichaetial bracts also need careful examination. 122 THE JOURNAL OF BOTANY Mrs. Britten from Mitten's herbarium was one, " Ceylon, Dr. Thwaites, 239 in part," which is different from any of the plants so far mentioned. It is dioicous, has leaves very similar to those of S. monoicum, but wider, more concave, with more abruptly narrowed and rather longer and finer points, the seta much longer, 2-5 cm., and the perichaetial bracts very distinct, the inner having a shortly loriform denticulate acumen, at the base of which the leaf is abruptly widened, with two or three coarse teeth at each side. It agrees, in fact, exactly with the plant described and figured in the Bry. Jav. as H. Gedeanuvi C. M. (ii. 208, t. 307), where the characteristic perichsetial bract is shown as described above. Assuming this to represent C. Miiller's species (as to which the authors express a certain amount of doubt), this speci- men of Thwaites's is certainly referable to S. Gedeanum. Obviously, therefore. Mitten had no very clear conception of his Ster. acuti- rameus, and Thwaites's C. M. 239 and 239 &, issued by Mitten under that name, contain at least three distinct plants. I have endeavoured to find out what the co-type of Ster. acutirameiLs Mitt, is, " In mont. Khasian., ad Moflong, in pinetis, Griffith ! " No specimen of this is to be found in either of the national collections, nor do any of Griffith's specimei:is exist in Mitten's own herbarium. But under S. Gedeanum at Kew there is a specimen determined by Mitten as " Acropormm Gedeanum C. MiilL, Khasia, Moflong, Herb. Griffith, 245," which I take to be in all probability Mitten's co-type of Ster. acutirame^is, but as to which he probably revised his opinion at a later date, naming it A. Gedeanum (otherwise he would without doubt have recorded Griffith's 245 in the Muse. I. Or. under S. Gedeanus, but the only record of that species there is " In Ceylon, Gardner ! "). I am not able to say certainly to what species this belongs ; it does not, however, agree with the H. Gedeanum of the Bry. Jav., either in leaves or in perichgetium ; nor with S. monoicum, though near it. I propose, therefore, to drop the name of S. acutirameum, as representing a composite species. It may be argued that the type- specimen being the plant described as H. monoicum in the Bry. Jav., the name acutirameum should be retained for that plant. The description and figures in the Bry. Jav., however, have established the species quite clearly, and as it is a widely distri- buted one, it would be unfortunate to disturb the nomenclature. Moreover, it is quite clear that Mitten had no definite idea of the species described as Ster. acutirameus — I have shown that three or four different species were so named by him at one time or another (and a Burmese plant, Moulmein, leg. Parish, 96, also determined by Mitten as Ster. acutirameus, is probably distinct from them all) — while there is almost a certainty that one of the two plants actually cited by him for Ster. acutirameus (Griffith's Khasian plant) was a different species from the actual type. Nor have I in the above remarks in any way attempted to exhaust the number of different species that have passed in one herbarium or another as S. acutirameum ! I believe, therefore, that there will MISCELLANEA BRYOLOGICA 123 be a consensus of opinion in putting a merciful end to this much tormented name. Daltonia nov^-zelandi^ Mitten. In Journ. Linn. Soc. (Bot.) iv. (1859), p. 95, Mitten described a plant under the above name from " New Zealand, ravines near Wellington, Stephenson." In the Handbook of the Neio Zealand Flora this is reduced, without comment, to a synonym of D. nervosa (H. f. & W.), and D. novce-zelandice forthwith disappears from bryological literature. Quite recently Mr. W. Gray, from whom I have received many interesting New Zealand mosses, sent me a packet labelled " 168, Daltonia mixed with Saulovia, Eopuaranga, Wairarapa, N. Z," On examination it proved to be a species unknown to me, and appeared to agree quite well with Mitten's original description of D. novce-zelandice. Further, it became clear that this description by no means fitted the plant now known as Bellia nervosa (H. f. & W.) Broth. {Hookeria nervosa H. f. & W.), a robust plant with stout nerve reaching to apex, and very short seta, which with D. straminea Mitt, has been placed in the new genus Bellia by Brotherus on account of these features, and especially of certain definite peristome characters. Now Mitten, among other things, has for his D. novce-zelandicB "nervo sub apice evanido," while he describes it as very closely resembling D. splachnoides in habit ; neither of which remarks applies at all to Bellia nervosa. I therefore attempted to see Stephenson's plant, of which Mitten writes that only a very small quantity had been seen. None, however, is to be found in the National Collections ; nor is there any plant in Mitten's own herbarium so labelled; but Mrs. Britton has found with D. pusilla H. f. & W. a mounted specimen, "Daltonia, New Zealand" — the only New Zealand Daltonia to be found in the collection. This is in all probability Stephenson's plant. It agrees exactly with W. Gray's 168, and with Mitten's original description of D. novce-zelandicB, and is an entirely good species, in no way closely related to Bellia nervosa (H. f. & W.). It is most nearly allied, perhaps, to the Tasmanian D. pusilla H. f. & W., but that is a smaller plant, with decidedly narrower and more attenuated leaves, and distinctly narrower cells.''' The areolation in D. novce-zelandice is rather markedly wide and not much incrassate, while the leaves are somewhat abruptly narrowed above to a moderately wide, not very attenuated nor very acute point. It is at present known only from the two localities above mentioned, in the North Island. Beachythecium trachypodium (Funck) B. & S. in Britain. Brachytheciuvi trachypodium is recorded in the Census Cata- logue of British Mosses (1907) for v.-c. 88, i. e., Mid Perth. The record depends, I believe, entirely upon a specimen in Mitten's * It is doubtfully distinct, as Fleischer suggests (Musci . . . vonBuitenzorg, iii. 960), from D. angiiMifolia Dz. & Mb. ; which again is scarcely separable from the probably widespread D. splachnoides Hook. & Tayl. 124 THE JOURNAL OF BOTANY herbarium, but I am not aware that it was ever pubHshed by Mitten as a British plant. The specimen has been kindly sent to me by Mrs. Britten. There are two sheets. One has been labelled at New York, " Found with Brachrjthecium trachypodium (Brid.)." It is a good fruiting specimen of trachyj^odium, with Mitten's sketches attached, no locality or date, the only data given being " Mr. Black" in Mitten's hand. It is pretty obvious that this is a Continental specimen placed side by side with the British plant for comparison. The other consists of a small packet, labelled in Mitten's hand " Hypnum tr achy Imodium '^. Encalypta rhahdocariM,-'' Ben Lawers, Hb. Hooker." It is accompanied by a drawing of a leaf and basal cells, which might very well represent B. trachypodium. The specimen itself is mainly the Encalypta, with one or two stems of a Brachythecium not unlike some of the less well characterized forms of B. trachypodiuvi, and with leaves that certainly show decided though weak striation — one of the principal characters separating B. trachypodium from B. velutinum. For all this I should consider it to belong to B. velutinum rather than to B. trachypodmm. The latter species is usually distinguished by its more robust, more rigid habit, with rather stout obtuse branches; the Ben Lawers plant is more silky, with rather slender, tapering branches. In habit and leaf form the Ben Lawers specimen is exactly in agreement with certain forms of B. velutinum, notably var. pralongum B. & S. (e. g. Husnot, Musci Gall. No. 741), to which var. I should refer it. I do not think the slight striation of the leaves, unusual as it is in B. velu- tinum, can be held to outweigh the other characters, and I should certainly consider it very unsafe to base a British record of B. trachypodium on this plant. The specimens in question are now placed, at Mrs. Britten's request, in the Kew Herbarium. INDEX SPECIES IN A FLOEA. By the Eev. E. Adrian Woodruffe-Peacock, F.L.S. What is an index species ? A plant that points out the in- coming of a new combination of circumstances, or the existence of such circumstances when hidden from view. A party of botanists were working a marsh dyke with its glory of Lythrum, Utricularia, Sparganium, Juncus, and Carices. Suddenly they came on Stachys palustris, intruding, as it were, amongst the true marsh species. The banks of the dyke were still firm peat, of the Sparganium, Juncus, and Carices formation, but the bed of the dyke had reached the Oxford Clay lying below at the spot, and this permitted S. palustris to survive and flourish. There was nothing very wonderful in being able to point out the junction spot in a second and this plant as its index species ; the wonder should be that such things are ever passed over, but it • This has been written over " Btreptocarpa,^' but the specimen is E. rhabdocarpa. INDEX SPECIES IN A FLORA 125 was looked on as astonishing by other workers because they had never given their minds to true observation." Surely true ecology (useful but hateful word !) is the art of observing the sequences in nature which follow from changed circumstances. Master the natural surroundings of the plants under all conditions, then the slightest change of circumstance even when invisible, as in the case given, will be clearly demon- strated to the mind. Other instances of invisible influences are easily given. The Chalky Boulder Clay lying in Lincolnshire west of its mother rock, the high Chalk of the Wolds, is not a fairly uniform bed of clay like the Oxford and Kimmeridge Clays. It is made up of layers growing more and more chalky till the basement layer is reached. The lowest zone may contain as much as 90 per cent, of carbonate of lime ; or, in other words, practi- cally be chalk moved to another situation. In working a Chalky Boulder Clay area of any considerable size, lying near its mother rock, it will be observed that where the bed feathers out to a thin edge I on the rock below, or where the streams which drain it have cut shallow valleys through its upper and more clayey layers into its lower and more chalky depths, the flora at once shows a change of species : in pasture by the appearance of Plantago media, in hedges of Sison Amomum, and generally by the coming in of Senecio erncifolms, and such species in ditches as Ranunculus auricomus, var. depauperata, or var. apetala. I have never found the typical plant under such circumstances. The Lower Lias Clays and Limestones give a similar set of varying circumstances, but caused in another way, and generally more visible. These beds were laid down in a fairly shallow sea- bottom not far from land, and consist of thick masses of stiff blue clay interstratified with beds of good building limestone, which, being more resisting to denudation than the clay, frequently make low escarpments. The flora at once indicates the change from clay to stone, though the outcropping zones of limestone are sometimes practically invisible and not more than a few yards across. Campanula glomerata appears on the narrowest zones, and C. Trachelium and C. latifoUa along with it on the wider ; and B. auricomus is found also in its most perfect state — the only * I noted at the time, for I was specially collecting insect visitors that day, that Buinbus agrorum — the only humble-bee that I have ever personally recorded as visiting Stachys j^alustris for honey, though Mr. G. ¥. Scott-Elliot has recorded three others — only flew as far as this spot on the dyke. It sucked honey, and then circled round and round, and finally returned by the way it had come, along the upland part of the dyke, quite omitting the marsh stretch below. I do not mean to imply by this that B. agronim is not a marshland species. It is common enough here on our peat carrs, often taken on Lythrmn ; in fact, the only species of Bombus I have ever taken personally on it, though Mr. Scott-EUict records two others. I simply wish it to be understood that the bees observed that day did not mix the honey taken from the upland flora at this junction spot with that from the true marsh species. This fact is, I take it, worth record. t This feather edge is beloved of certain plants, and also varieties of Helix virgata, which are not found on the more clayey masses above. 126 THE JOURNAL OF BOTANY place where I have found the typical form. These and many others are all good index species in working such a rock in our Lincolnshire area. It must be remembered that index species may be quite local. Wide field-work demonstrates that they may be peculiar to a very limited area indeed as true index species. This will be perfectly proved when the true ecological method has been applied widely. For instance, Potentilla argentea is the best index species for the Spilsby Sandstone in Lincoln- shire. It is confined in this county solely to that bed ; but beyond our area it is a mere sand index species. Not only do index species proclaim what exists now under given circumstances, but they can be safely used to demonstrate in limited areas what once existed, though the circumstances have now completely changed. This is one of the great advantages of intensive ecological study of limited areas when the results are applied to solving the problems of county or vice-county floras. To give an instance of this, take the Pyrola. They are recog- nised as woodland or woodland scrub species." In Lincolnshire some most curious facts come out of a historical consideration of the position in the past of our only species, P. minor. The great block of parishes for which it is recorded lies on the eolian sands at the foot of the escarpment of the wolds for a distance of eight miles north and east of Market Easen. Now on these sands Pimcs sylvestris grew as a self-sown species from prehistoric times till about 1840, if not later. Beyond this area Pyrola is found in a few isolated spots. These places are worth careful study to see whether it is a good and safe index species of ancient, but now departed, pinesques in this county. In 1840 it was recorded, under the mistaken name of P. rotundifolia, for Laughton Common.! This common was the centre of a vast pinesque which flourished into historic times, and stretched north to south for fifteen miles over the eolian sand dunes of the Trent valley. I have personally seen tons of pine which have been dug out of the peat in various parts of this old forest. One spot was called Welfholme — a truly suggestive name for part of the forest that once was there. P. media has been destroyed in this area since 1840 by allowing rabbits to increase beyond all reason. | * The Marine Sand Dune variety arenaria of P. rotundifolia is no excep- tion to the general rule ; where I have seen it growing in the valleys of the Lancashire eolian hills, the ground was scrubby with Halix repens, &c. f There can be no question as to this, for the County Herbarium possesses a specimen of P. minor from Nottingham named P. rotundifolia by the same authority, the Rev. J. K. Miller, a very good botanist for his time (1787-1855). \ In the Journal of the Ecological Society, i. p. 273, some doubt is cast on the point whether hungry rabbits will eat Senecio Jacohcea. I can only say round the spot where P. minor formerly grew — for it is known exactly — S. Jacobcea was badly eaten by them. The trees of a fagesque of forty years' growth were badly barked, and many plants were locally exterminated along with Pyrola. The most astonishing fact I discovered in this rabbit inquiry \fa,sihe\ynyAnthriscusi'ulfia)is met and adapted itself to this rodent's appe- tite. The plants were but four or five inches long, buried in moss completely out of sight with the exception of their flowers and seeds. CASUAL PLANTS IN MIDDLESEX 127 The next area we come to for which Pyrola has been recorded is still in N. Lines. (54), between Lincoln and Boston. The Kirkby moor and Koughton bed of Plateau Gravel and Coningsby bed of Old River Gravel are still its home. There I have proof that the last of the self-sown prehistoric pinesque scrub was not finally uprooted and destroyed till past the middle of last century. The only other recorded spot for this species is in S. Lines. (53) in a wood on the same Old River Gravel that the ancient pinesques frequented at Coningsby. There is no peat in this locality, as there is in all the others, to preserve for us proof that pinesques formerly existed in this spot ; but considering what is known of the other districts, it is surely safe to say that they were once there too. I can prove their presence in the immediate neigh- bourhood on the same Old River Gravel. True ecology will not rest there ; it will explain much lying in my notes which is as yet hidden from the wisest. There are index species, varieties, and even hybrids ; and we do not yet know to what they point. For instance, the Rev. H. J. Riddels- dell borrowed the arranged County Herbarium series of Ajmmi nodifloruvi and A. inundatum, and their forms. The first is wide- spread and common, growing in all kinds of waters ; the second, thinly but widely scattered, and found in neutral or acid waters. A form found with us was returned named Moorei Riddelsdell, which the late Canon Fowler once suggested in conversation was nothing but a hybrid between them, on account of its mixed characteristics. The curious thing is that all the specimens re- turned named Moorei {i.e., approximately inundatumx nodifloruvi) were taken in the known inundatuvi conditions — distinctly acid waters ; in the Trent valley too, with the exception of a specimen I have heard of, collected by Mr. G. C. Druce at James Deeping parish on the very borders of this county. I have no notes of any insects visiting the flowers of these Apiums, neither has Mr. Scott- Elliot, which seems extraordinary — not even a Thysanopteron. CASUAL PLANTS IN MIDDLESEX. By J. E. Cooper. The following list, which is far from being exhaustive, may serve to show the great variety of casual and alien plants to be found on waste ground and building land round the Metropolis. All the plants mentioned were collected by the writer. The list excludes (1) plants which are apparently native and (2) those which are presumably garden escapes. The following abbreviations are used : — C. E. = Crouch End. H. M. = Hackney Marshes. E. F. = East Finchley. M. Hill = Muswell Hill. F. = Finchley. Y. = Yiewsley. For the identification of several plants the writer is indebted to the Director of the Royal Gardens, Kew\ He also desires to 128 THE JOURNAL OF BOTANY thank Messrs. E. G. Baker, S. T. Dunn, and A. B. Jackson for assistance in naming others. Kanunculace^. — Banunculus arvensis L. E. F. 1883 ; H. M. 1912. — R. sardous Crantz. Uxbridge, 1910. — Delphinium Ajacis L. Y. 1910. PAPAVERACEiE. — Rc&meria hybricla DC. Near Potter's Bar, 1912. Crucifer^. — Alyssum incanum L. C. E. 1897 ; F. 1900, 1909 ; Y. 1910-11 ; M. Hill, 1904 ; Cowley, 1913. — SisymhrinmiKinnonicum Jacq. E. F. 1909 ; M. Hill, 1902 ; F. 1910 ; H. M. 1910-13 ; Y. 1909-10. — S. ColumncB Jacq. H. M. 1909, 1912-13. — S. Loeselii L. Hampstead, 1910 ; Y. 1911-12. — S. SopJiia L. Y. 1909-13 ; F. 1910 ; H. M. 1912.— Erysimum cheiranthoides L. E. F. 1882, 1909; Y. 1909-10, 1912-13. — ^. orientale Mill. E. F. 1907; F. 1909; H. M. 1913. — ^. repandum L. Y. 19Q1 . — Camelina sativa Crantz. C. E. 1887 ; M. Hill, 1906, 1909 ; Y. 1908-9, 1912-13.— Bras sica nigra Boiss. H. M. 1912.— S. elongata Ehrh. F. 1911-12. — 5. alba Boiss. M. Hill, 1912; H. M. 1912-13.— Diplotaxis muralis DC. F. 1907-8; Y. 1908-13. — Lejndium Draba L. M. Hill, 1906, 1912-13 ; H. M. 1909 ; Hanwell, 1910, 1912 ; Y. 1910, 1912-13.— L. ruderale L. Waste ground every- where.— L. virginicum h. M.Hill, 1902. — Thlaspi arvense h. C. E. 1887; M. Hill, 1892, 1902; F. 1909; Harefield, 1910; H. M. 1913.— Neslia paniculata Desv. C. E. 1899 ; E. F. 1907 ; H. M. 1912-13.— Bapistrum orientale DC. Y. 1912-13 ; H. M. 1912-13.— B. per enne DC. Y. 1910; F. 1912.— E. rugosum Berg. H. M. 1913.— Euclidiimi syriacum Br. H. M. 1912. Caryophyllace^. — Dianthus Armeria L. E. F. 1883. — Saponaria Vaccaria L. C. E. 1885; E. F. 1907; H. M. 1912-13; Y. 1913. — Gypsophila elegans Bieb. H. M. 1913. — Silene conoidea L. H. M. 1912.— >S. anglica L. Y. 1908 ; F. 1909-10; H. M. 1913.— S. nutans L. H. M. 1912.— S. noctiflora L. C. E. 1897; M. Hill, 1902; H. M. 1910; Y. 1909, 1913. — Lychnis Githago Scop. F. 1908-9 ; Y. 1909, 1913. Malvace^. — Hibiscus Trionum L. Fortis Green, 1909. — Malva pusilla^m. Y. 1909-10. — M. parviflora Jj. Fortis Green, 1911 ; Y. 1912 ; H. M. 1912. GERANiACEiE. — Geranium pusHlum'BviXiTi. E. F. 1908. — Oxalis corniculata L. Muswell Hill Road, 1897. — Impatiens p)arviflora DC. Hampstead, 1891, 1908 ; C. E. 1897 ; Uxbridge, 1910. Leguminos^. — Tngonella Foennm-gracum L. Y. 1908, 1913._r. ccBTulea Ser. H. M. 1912-13.— Medicago falcata L. M. Hill, 1904; H. M. 1912.— ill. denticulata Willd. Y. 1908-9; H. M. 1912-13.— Var. apiculata Willd. Y. 1909 ; H. M. 1910, 1912.— -M". lappacea Desv. Y. 1908; H. M. 1912.— Melilotus alba Desv. Waste ground, many localities. — M. officinalis Lam. F. 1906; Y. 1908, 1910; E. F. 1909; H. M. 1909.— M. indica All. C. E. 1898; M. Hill, 1902; F. 1908, 1910; Y. 1909-10, 1913; H. M. 1909-10. — Tr if olium lappaceim L. CASUAL PLANTS IN MIDDLESEX 129 E. F. 1907 ; Y. 1908 ; F. 1910.— T. ochroleucon Huds. F. 1907 ; M. Hill, 1907. — r. arvense L. F. 1908; B. F. 1909.— T. imrviflorum Ehrh. H. M. 1913. — T. resupinatum L. M. Hill, 1902. — Anthyllis Vulneraria L. Eoadside near Staines Moor, 1910-13. — Scorpiurus subvillosus L. Fortis Green, 1911. — Vicia bum L. Y. 1912 ; H. M. 1912.— F. villosa Eoth. Y. 1908, 1913. E. F. 1908.— F. pseudocracca Bert. C. E. 1897 ; F. 1909 ; Y. 1911-13 ; H. M. 1912-13. — F. 2)eregrina L. H. M. 1912.— Lathijrus Aphaca L. E. F. 1909 ; H. M. 1909, 1912 ; Y. 1912.— L. hirsutus L. C. E. 1897; E. F. 1909; H. M. 1912-13.— L. Cicera L. H. M. 1912. 'RosACEM.—PotentiUa hirkt L. C. E. 1897 ; M. Hill, 1902 ; Y. 1908-9 ; F. 1910.— P. argentea L. Highgate, 1887-8 ; Y. 1901. Umbellifee^. — Coninm viaculatum L. Y. 1912. — Bupleurum rotundifolium L. Whetstone, 1906. Y, 1908. — B. j)rotractum L. & H. H. M. 1913.— Cantm Carvi L. Highgate, 1887 ; E. F. 1909 ; Y. 1910-11, 1913. — Fceniculum vulgare Mill. H. M. 1912-13. — Gonandmm sativum L. H. M. 1913. — Caucalis daucoides L. H. Mill, 1906 ; H. M. 1912. — C. arvensis Huds. Harefield, 1913 ; near Colnbrook, 1913. Kubiaceje. — Galium tricorne Stokes. Highgate, 1887 ; F. 1908; Y. 1910, 1913; H. M. 1912-13. — A sjm-ula arvensis L. E. F. 1908. Valerianace^. — Valerianella dentata Poll. Harefield, 1912. DiPSACE^. — Dipsacus fullonum L. Y. 1908, 1910-11, 1913. Composite. — Erigeron canadenseh. Highgate, 1897; M.Hill, 1905; F. 1906, 1908; Y. 1909; H.M. 1909-10.— ^. acre L. Y. 1910; Harlington, 1910.— Ambrosia trifida L. Highgate, 1897 : H. M. 1912.—^. artemisifolia L. F. 1900; M. Hill, 1902.— Xa7ithmm spinosum L. H. M. 1913. — Achillea tanacetifolia All. Y. 1910. — Guizotia ahyssinica Cass. Y. 1913. — Anthemis tinctoria L. Y. 1910 ; F. 1910.— .4. Cotula L. F. 1910; Y. 1912 ; H. M. 1912. —A. arvensis L. F. 1907, 1910; Y. 1910, 1912 ; H. M. 1912.— Matricaria suaveolens Buch. Highgate, 1907, 1910 ; F. 1908 ; H. M. 1912-13.— Cotula coronopifolia L. M. H. 1913.— Artemisia biennis ^im. H. M. 1913.— .4. longifolia Mitt. Y. 1913.— Senecio viscosus L. Highgate, 1897 ; M. Hill, 1900 ; Y. 1909, 1913; F. 1910; H. M. 1909-10.— Car dims pycnocephalus L. {tenuiflorus Curt.). Y. 1913. — Cnicus eriophorus Scop. C. E. 1896. — Silybum Marianum Gaertn. E. F. 1907-8, 1910; Y. 1908-9, 1913. — Centaurca calcitrapa L. Fortis Green, 1911. — C. solstitialis L. Y. 1912 ; Harefield, 1912. — C. melitcnsis L. Y. 1912 ; H. M. 1912-13.— Car^/ia?/iws tinctorius L. E. F. 1906 ; Y. 190Q-9. —Picris echioides L. E. F. 1883; F. 1910; Harefield, 1912.— Lac^wca virosa L. Y. 1908-13 ; M. Hill, 1910 ; Harefield, 1913.— L. Serriola L. Y. 1913. Primulaceje. — Anagallis foemina Mill. Highgate, 1911. BoRAGiNACE^. — Lappula echinata Gilib. Y. 1910; H. M. 1913.— 5ora^o officinalis L. E. F. 1908; Y. 1913 . — Anclmsa Journal of Botany. — Vol. 52. [May, 1914.] l 130 THE JOURNAL OP BOTANY officinalis L. E. F. 1908. — Lithospermu7n officinale L. Y. 1910-12 ; H. M. l9lS.~Echium vulgare L. M. Hill, 1903 ; Whetstone, 1906 ; F. 1909 ; E. F. 1909. SoLANACE^. — Atropa Belladonna L. C. E. 1896; Harefield, 1908. — Datura Stramonium L. var. tatula L. Y. 1913. — Hyoscyamus niger L. Neasden, 1908; Y. 1910; F. 1909-10; H.M. 1910, 1912-13. ScROPHULAEiACEiE. — Autivrhinufn Orontiiim'Lt. Stan well Moor, 1913. — Alonsoa peduncularis Wetts. F. 1909. ljKBih.T!M.— Salvia verticillata L. M. Hill, 1905 ; F. 1907-8 ; Y. l^Qd.—Marruhium vulgare L. M. Hill, 1905 ; Y. 1911 ; H. M. 1912. — Stachys arvensis L. Stanwell Moor, 1913. — S. annua L. C. E. 1897 ; F. 1910 ; Y. 1912. — Galeopsis angustifoUa Ehrh. M. Hill, 1900 ; F. 1910 ; Y. 1910, 1912.— G. Tetrahit L. Highgate, 1883 ; E. F. 1908, 1910 ; H. M. 1909 ; F. 1910. — Leommis Cardiaca L. Hendon, 1907. Plantaginace^. — Platitago arenaria W. & K. C. E. 1896; H. M. 1912. AMARANTACEiE. — Amarautlius retroflexus L. E. F. 1897 ; F. 1908, 1910-11; Y. 1908-11, 1913; H. M. 1912-13.—^. Blitumh. H. M. 1913. GnEiio-po-DiAC-EM.—ChenojJodium murale L. M. Hill, 1896; E. F. 1906.— C. Tubrum L. E. F., F., Y., H. M. (very abundant). — C. polyspermum L. E. F., F., Y., M. Hill (abundant). — C. glaucum L. M. Hill, 1906; Y. 1909.— C. Bonus-Henricus L. F. 1912-13. — C. amhrosioides L. Y. 1911. — Suceda mariiima Dum. H. M. 1909. PoLYGONACE^. — Bumcx limosus Thuill. Hornsey, 1887-8; C. E. 1897 ; F. 1911 ; H. M. 1913. EuPHORBiACEJE. — MercuHalis annua L. E. F. 1909 ; H. M. 1909-10, 1912; Y. 1913. XjRTicACEiE. — Cannabis sativa L. C. E. 1897 ; Y. 1908-9. Graminaceje. — Panicwn Crus-galU L. Highgate, 1896 ; C. E. 1897; M. Hill, 1902; F. 1908; Y. 1908-10; H. M. 1912-13.— P. miliaceum L. C. E., F., E. F., Y. (abundant). — P. capillare L. C. E. 1897 ; Highgate, 1899.— Setaria viridis Beauv. C. E. 1897 ; M. Hill, E. F., Y., F. (often abundant). — S. glauca Beauv. (occurs with the last in all localities). — Phalaris canaricnsis L. (abundant at times in all localities). — P. minor Retz. H. M. 1912. — P. para- doxa L. E. F. 1907 ; F. 1908 ; Y. 1908, 1912 ; H. M. 1912.— P. angusta Nees. H. M. 1913. — Anthoxantlmm aristatum Boiss. H.M. 1909, 1913; Y. 191^.— Agrostis scaira Willd. F. 1910; H. M. 1912; Y. 191^.— Apera Spica-venti Beauv. M. Hill, 1906; E. F. 1909; H. M. 1909; Y. 1910-13. — Air a cai-yoj^hy Ilea L. H. M. 1913. — Trisetum paniceum Pers. H. M. 1913. — Avena fatua L. E. F. 1908-9 ; F. 1908-10 ; H. M. 1909. — Cynosurus echinatush. E. F. 1910, F. 1910-11; Harefield, 1910 ; H.M. 1913.— Glyceria distans Wahlb. H. M. 1909-10, 1912 ; B. F. 1910, — Festuca myurus L. Y. 1912-13; H. M. 1913. — Bromus BRITISH PLANTS 131 rigidus Eoth. M. Hill, 1907; Y. 1911-13; H. M. 1913. — 5. tcctorum L. Potter's Bar, 1912; H. M. 1913. — 5. secalinus L. F. 1913. — B. arvensis L. E. F., M. Hill, Y. (sometimes ohnn- diint).—B.brizceformisW. H. M. 1912 ; Y. 1913.— B. unioloichs H. B.&K. E.F. 1906-8; F. 1908-10; Y. 1909, 1913; H. M. 1913. — ^gilops cylindrica L. Fortis Green, 1909. — ^. triuncialis L. H. M. 1912.~H:ordeum jubatum L. H. M. 1909-10. REVIEWS. British Plants. The Cambridge British Flora. By C. E. Moss, D.Sc, F.L.S., assisted by Specialists in certain Genera : illustrated from Drawings by E. W. Hunnybun. Volume II. Salicacece to Chenopodiacecs. Folio ; paper boards, pp. xx, 206 ; 206 plates. Price £2 5s. net. Cambridge : University Press. British Floivering Plants. Illustrated by Three Hundred full-page coloured plates [by Mrs. Henry Perrin] , with detailed descriptive Notes and an Introduction by Professor Boulger, F.L.S. 4to, buckram gilt, pp. xlv ; Ixvi plates with text. London : Quaritch. It is somewhat remarkable that there should appear within a few days of each other two works devoted to the British Flora which, from their different standpoints, may be regarded among the most important of their class. We must go back to the beginning of Syme's edition of English Botany half a century ago for anything equalling in importance The Cambridge British Flora, while we shall find nothing to compare in sumptuousness of get- up with the new work on British Floioering Plants. The appearance of Dr. Moss's work — we note that, originally appearing as editor, his name now stands as author — has been anticipated by British botanists with the greatest interest : not only to them does it appeal, for its completeness and attention to detail entitle it to take rank among works of Continental impor- tance. The Cambridge University Press has been fortunate in securing the services of Dr. Moss, than whom no one more com- petent for the task could be found. By a combination as admirable as it is rare. Dr. Moss is at once an acute field botanist, a diligent investigator of herbaria, and a student of botanical literature : in a comparatively short time he has attained a leading position among British botanists and has acquired a knowledge of the history of his subject equalled by few. Mr. Hunnybun's drawings are all made from living plants, so that the work may be regarded as representing more fully than has been hitherto done our knowledge of British botany at the present day. The many features new to British botany which the book contains begin with the arrangement, which is that of Engler's Syllabus — not hitherto adopted in any British flora. It has been found convenient to begin with the second volume, which con- tains orders of unattractive appearance though of great botanical interest : a list of these was given so recently in these pages 132 THE JOURNAL OF BOTANY (p. 87) that it is unnecessary to repeat it here. The treatment of the species is so detailed that only one critically versed in their study could offer any useful comment upon it. We must therefore content ourselves w^ith some remarks upon the general plan and execution of the work, as detailed by Dr. Moss in his introduction. The objects of the work are thus stated : " First, an attempt is made to register the present state of knowledge with regard to British plants — their classification, their names, their characters, and their distribution. Secondly, an attempt is made to relate British plants to the allied forms of foreign countries. And thirdly, a hope is entertained that the work will result in stimu- lating further research concerning British plants, particularly with regard to the study of their variations and the distribution of the less well-known forms." The section on nomenclature is interesting and clearly set forth, but we regret that the Eules and recommendations laid down by the International Congress at Vienna have not been altogether, instead of "in general," adopted. We are entirely at one with Dr. Moss in rejecting the accidental binominals of Hill and other pre-Linnean books, though we are not so clear as to the exclusion of Adanson as "pre-Linnean in character although not in chronology " ; but we think the names of species ought, in all possible cases, to stand as in the first edition of Linnaeus's Sjjecies Plantarum. What is referred to as "the Kew rule" was not happily so named ; for although the retention of the earliest trivial name received by a species when placed under its correct genus — always observed by the British Museum and usually by French botanists — was often followed at Kew, it was there sub- ordinated to " convenience " when such subordination was con- sidered desirable. Our chief objection however is to the use of small letters for trivial names, as to which Dr. Moss expressed his views in this Journal for 1913, p. 21 : as we then said, " it seems to us undesir- able to depart from a practice which is sanctioned both by rule and custom," and we cannot agree with him when he says that no "precise rule or custom" exists. The matter is hardly of sufficient importance to make a fuss about, and we allow Dr. Moss — who, like the rest of us, likes to have his own way and is perhaps more fortunate because more insistent in getting it — to follow his plan in our pages rather than deprive our readers of his valuable contributions. But it seems undesirable to depart from established custom unless for some considerable advantage ; and it is not without significance that the Kew botanists, who at one time adopted the practice, soon returned to the general custom. A similar departure from general use is noticeable in the printing of the name of the species at the head of each description without any appended authority : Dr. Moss gives no reason for this method, which was opposed by all the botanists who spoke at the meeting held to consider the plan of the Flora. The synonymy given for each species includes certain names BRITISH PLANTS 133 taken from pre-Linnean British authors, chiefly from Gerard and Eay — a beautiful photogravure portrait of the latter forming the frontispiece to the volume — followed by a number of references to the more important works in which the plant has been described, under the name adopted or under others : to each reference the date is appended. No attempt is made to include folk-names, although one English synonym is given for each species when such name is generally known. A good deal of attention has been given to distribution, which in many cases is illustrated by useful maps. In the matter of classification, the subdivision of orders is carried out very thoroughly ; many of the subdivisions are new : thus of the three subclasses of Engler's Archichlamydeae, two are here first estabhshed, and Populus is grouped under four series, all of them new. Although the numerous subdivisions of species which make some Continental books practically unworkable is comparatively restrained, there are amply sufficient here to tax the observation of the collector ; thus of Atriplex imtula four varieties and three forms of one of these are described : the reference appended to the authority for some of the names has a somewhat strange appearance — one does not at first sight recog- nize that " var. crecta forma crassa Moss and Wilmott in Camb. Brit. Fl. ii. 174 " actually refers to the page in the present volume on which the name appears. The book is handsomely printed, but might have been better arrayed. It would perhaps be too much to expect that a new page would be begun for each species, but certainly each family should start on a fresh one. It is not easy to understand why the genera should be printed in small black type while the species are in large capitals. The different types are however on the whole judiciously employed ; and the use of a large quantity of small but clear type allows the inclusion of a vast amount of information. A few names will come to most botanists as new : Pojnilus taca- mahacca Miller (1768) supersedes P. candicans Aiton (1789) ; Oxyria appears as Bheum digynum Wahlenberg. Mesemhryanthe'nnim edule occurs for the first time in a British flora, being "naturalized near the sea on cliffs, rocks, old walls, and hedgebanks in the Channel Islands, Cornwall (including the Scilly Isles) and in the Isle of Wight"; Quercus Ilex and Q. Cerris are also regarded as naturalized in southern England ; Salsola Tragus, " not indi- genous," is included as having been found in various localities. The genus Salsola, by the way, is contributed by Mr. C. E. Salmon; the Rev. E. Marshall has undertaken Betula, a genus at which he has long worked ; the rest of the book is by Dr. Moss, with the assistance of Mr. Wilmott in Atriplex and of Dr. E. J. SaUsbury in Salicornia. Turning to Mr. Hunnybun's plates, it is to be noted that "each plant or portion selected has been drawn natural size," and is " reproduced without reduction or enlargement " : "each drawing has been made from a fresh plant, the name of which has been vouched for by some competent authority whose letter of identifi- cation " — and we assume also the specimen — " is preserved in the 134 THE JOURNAL OF BOTANY Cambridge University Herbarium." In a few cases, cultivated examples have been used, but by the co-operation of numerous botanists Mr. Hunnybun has been supplied with specimens of most of the species which were not gathered by himself. The orginal pen-and-ink drawings were presented by Mr. Hunnybun to the University ; many wei'e previously circulated among botanists, who expressed a high opinion of their artistic merit and scientific value. The number of forms presented is remarkable and, at any rate so far as British botany is concerned, unique ; thus Pojmlus iind A triplex are represented by 17 plates, Ulmus by 16, Salicornia by 12. They are reproduced by a special process and are mostly in outline ; the few dissections added in most cases, are not, we understand, considered by experts as altogether satisfactory. For the botanical accuracy of the plates the name of Dr. Moss is sufficient guarantee, and Mr. Hunnybun is to be congratulated on the success with which he has, in most cases, succeeded in conveying the habit — the port, as the French more expressively put it — of the plants : this is especially notable in the PolygonacecB and Chenoj^odiacecB. The plates of Salicornia form an important contribution to the knowledge of the genus and will we think be welcomed by Continental as well as by British botanists. The branches of trees please us less ; from the artistic standpoint they leave a good deal to be desired. In many cases the specimens seem to have been thrown down any- how and to have been drawn as they fell — we do not suppose that such was the case, but it is certainly the impression conveyed. This is the more to be regretted because the size of the page enables the specimens to be fully displayed, and there thus seems no reason why they should be placed across one another as they are in numerous instances — e. g. nos. 2, 19, 20, 28 ; the object may have been to avoid a formal and diagrammatic appearance, but we think anyone who will contrast these with nos. 18, 22, 23, 26, will prefer the more formal arrangement. In many of the Atriplexes and Chenopods the separate leaves are scattered about in a casual way as if they had fallen from a height on to the paper, usually pointing downwards. In many cases, too, where a fragment of a plant only is given, the whole might well have been displayed. A study of the figures in some of the sixteenth-century herbals, notably those of Brunfels (1530) and Leonard Fuchs (1542-3) would, we think, have resulted in the production of figures not less accurate but far more artistic as well as more informing ; a comparison of Mr. Hunnybun's plate of Knotgrass with Fuchs's figure will illustrate our meaning. A like criticism applies to many of the plates which form the raison d'etre of Mrs. Perrin's handsome volume — the first of four — on British Floivering Plants, on the title-page of which, by an excess of modesty, her name does not appear. Many of these are extremely good, although, well reproduced as they are, they are not as good as the original drawings, in the exhibition of which we noticed some — notably that of the Sloe, to appear in a later volume — of really supreme excellence. Turning over the pages, BRITISH PLANTS 135 we should place among the figures of first rank those of the Scotch Fir, the Water Plantain, the Arum, the Fritillary, the Twayblade and the Eagged Eobin : all these are excellent. But the very accuracy of Mrs. Perrin's reproduction of individual specimens has in many cases resulted in an inadequate represen- tation of the plant presented. Even Mr. Hunnybun's plates occasionally suffer from this individuality, but we are inclined to think that in his case a wide general knowledge has enabled him to interpret the individual in the light of the species, whereas Mrs. Perrin seems to have restricted herself absolutely to what was actually before her ; nor has she always selected characteristic specimens. x\s examples of this may be mentioned the otherwise excellent figures of the Bur-reed, the flowers of which are too young to show the " golden tufts of ripe stamens," the Arrowhead, and the EjnjMctis called latifolia: this last, so far as it goes, could hardly be better, but almost all the flowers are unexpanded. The Bog Asphodel is somewhat past its prime ; the example of Orchis mascula does the species scant justice ; the figure of the Corn Cockle hardly adequately represents a very beautiful plant, either in the colour of both flower and leaves (many of the greens throughout are susceptible of improvement) or in the size of the former. The representations, seeing how much space was at the artist's disposal, are sometimes disappointingly inadequate; the Sheep's Sorrel, for example, might well have shown the under- ground growth which makes the plant so terrible a nuisance in gardens where it has attained a hold : the figure in Curtis's Flora Londinensis may be contrasted with this. The only really un- satisfactory plate in the book is that of the Stitchworts ; this according to the list contains three, but according to the text four species ; we are inclined to think the former correct, as we can find nothing in the least resembling S. palustris, either in flowers or foliage. Mr. Boulger's introduction and " descriptive notes " are of course accurate. The arrangement followed is that of Engler ; admitting that this is " the best linear grouping as yet achieved," we have doubts as to the wisdom of adopting it in a popular volume ; many folk who have no claim to be considered botanists have a sort of general notion that things begin with Banunculacea. To the scientific botanist the book hardly appeals — it contains only 290 species and entirely omits grasses and sedges, as well as " the less attractive water-plants " , hence we rather regret the care and cost which must have attended the production of the coloured analytical plates. A little more botany might we think have been added ; thus, when describing a species something might have been said about its allies — e. g. under S'parganium erectum some indication might have been given of the characters of S. simplex. The letterpress is an excellent example of a successful combination of science and popular lore, thus differing in this respect from most popular books. The English names which head the description are occasionally unfamihar — e. g. "Good Friday Grass," a purely local name for Luzula campcstris — 136 THE JOURNAL OP BOTANY and even inappropriate — anything less suggestive of " Spring Beauty" than Glaytonia perfoliata it would be difficult to con- ceive, however appropriate the name may be to the species in the same genus with which it is usually associated. It remains to be said that the book is beautifully printed and handsomely bound in buckram ; the colour-printing is on the whole excellent. BOOK-NOTES, NEWS, dc. At the meeting of the Linnean Society on April 2, Mr. Clement Keid showed a lantern-slide of photographs from seeds of a new species of Corcvia [C. intermedia) from the Pliocene Cromer Forest Bed. The same had also been found in a similar deposit in the Netherlands, at Tegelen. The plant forms the subject of a paper by Mr. Eeid in our present issue. Mr. R. Allen Eolfe exhibited a series of coloured drawings of five hybrid Ophryses, raised by M. Fernand Denis, Balaruc-les-Bains, France, from Ophrys ten- thredinifera Willd. crossed with the pollen of 0. aranifera Huds. ; together with the two parents. This was believed to be the first hybrid Ophrys raised artificially, and it proved the origin of a natural hybrid that has been recorded from three localities in Italy, and is known under the names of 0. Grampinii Cortesi and 0. etrusca Asch. & Grabn. The hybrids varied somewhat between themselves, but all showed an unmistakable combination of the characters of the two parents, particularly in the colour and markings of the lip, and in the peculiar combination of rose and green in the sepals and petals. M. Denis has a batch of some forty seedlings in flower or bud. At least eighteen natural hybrid Ophryses have been recorded, and Mr. Eolfe believed there were others. "After several years of preparation and discussion the Federal Government has just decided to create a large reserve, on the lines of the American Yellowstone Park, in the Lower Engadine for the protection and preservation of Swiss fauna and flora, especially the former. A subsidy of £1200 a year has been granted to the communes interested for a period of ninety-nine years, but the contract must be renewed every twenty-five years for the upkeep of the park, supervision, &c. The ' park ' is ready-made by Nature, for it is situated in one of the most lonely and most un- touched corners of Switzerland, containing mountains, forests, streams, and pastures which have been rarely visited except by smugglers who ' trade ' with Italy in contraband goods. The little village of Zernetz will be the headquarters of the park, through which there are only a few bridle paths, although it is not very far from the fashionable resort of St. Moritz. The reserve will be stocked by the authorities and at the expense of private societies. Owing to the lack of legislation the fauna of the Alps during the last fifty years has been almost exterminated, and the new measure will be only just in time to save several species." — Standard, April 2. 137 NOTES UPON TEESDALE PLANTS. By C. E. Salmon, F.L.S. The following notes have been put together mainly for the purpose of comparing the condition of the more uncommon Tees- dale plants in 1892, an ordinary English season ("three fine clays and a thunderstorm "), and in the phenomenally hot and dry summer of 1911. The first visit was from June llth-20th, when my brother and I had the advantage of the company of Messrs. J. B. and A. J. Crosfield, without whose guidance many of the " Teesdalians " would have been missed ; the second trip was from July 4th-20th, in the company of Dr. A. H. Fardon. It will be seen that, although the conditions may be pleasant for botanizing, a hot dry season by no means produces such a good display of mountain or rock plants as a normal summer. The Globe-flower had suffered extremely ; a meadow near High Force, yellow with the blossoms in 1892, did not show a single flower or fruiting-spike in 1911. D. = Durham, v.-c. 66 ; W. = Westmoreland, v.-c. 69. All the v.-c. 64 records were noted in 1911. Supposed new records have an asterisk. The Eev. E. S. Marshall, Eev. E. F. Linton, Messrs. H. and J. Groves {Characece), and Mr. A. Bennett have kindly helped me in naming critical forms. Thalictrum minus L. Limestone W. of Selside, v.-c. 64. Meconopsis cambrica Vig. Pot hole, near Selside, v.-c. 64. Draha incana L. Forest-in-Teesdale and limestone ridge near Ettersgill Beck, D., 1911. Cochlearia alpina Wats. By Harwood Beck, D., 1911. Lime- stone hills above Brough, W., 1911. Helianthevmm canum Baumg, var. vineale (Pers.). Flowering in profusion in its well-known locality in 1892 ; in 1911 not putting up nearly so many blossoming shoots. The Teesdale plant seems to have leaves much more glabrous above than those of the Great Orme's Head form, indeed some of them are quite hairless. Mr. Williams (Prod. Fl. Brit, pars x, 1912, 573) fails to distinguish vineale on "comparing examples from Clare with examples from the Welsh coast," but, as far as the Teesdale plant is concerned (I do not know the Irish), it seems worth separating at least as a " forma." Polygala amara L. In fair quantity in 1892 and 1911 ; the blue-flowered plant seen in 1911 does not grow intermixed, apparently, with the more frequent (as regards Teesdale) pink form. J. D. Hooker had evidently not seen the blue form in Teesdale; he says (Stud. Fl. 1884, 51): "The Teesdale form (P. uliginosa Fries) is rather more fleshy and has rosy flowers ; the Kent form (P. austriaca Crantz) is blue-flowered. I find no difference between their capsules. It is certainly the P. amara of Linn. Herb." Journal of Botany. — Vol. 52. [June, 1914.] m 138 THE JOURNAL OF BOTANY P. oxyptera Eeichb. Holwick Scars, 1911, v.-c. 65. Arenaria uliginosa Schleich. In good quantity in its well- know locality in both 1892 and 1911. A. gothica Fr. Seen in three stations in the Ingleborongh district, v.-c. 64. Sacjina nodosa Fenzl. E. side of Ingleborough, v.-c. 64. ■•'•Montia laviprosijerma Cham. Holwick Fell, 1911, v.-c. 65. Geranium pratense L. Near Malham village, v.-c. 64, a beautiful form occurred with pale lilac flowers. G. luciclum L. Near Great Musgrave and laneside near Hill- beck, Brough, W., 1911. Vicia sylvatica L. Swindale Beck, Brough, W., 1892. Dryas octopetala L. Plentifully in flower in 1892 in its only Teesdale station ; not a single blossom showing in 1911, although leaves were in abundance. Geuvi rivale x 2irhanum {intermedium Ehrh.). Barnard Castle, with the parents, D., 1892. Alchemilla alpestris Schmidt. Forest-in-Teesdale and High Force, D., 1911. -Near Cronkley Fell, v.-c. 65, 1911. '■'■"A. minor Huds. Forest-in-Teesdale, D., 1911. Mickle Fell, v.-c. 65, 1911. Saxifraga nivalis L. In its Westmoreland station in fair quantity in 1892 ; we failed to hit the right spot in 1911, but my friend Mr. A. Wallis tells me he saw it in 1913 (see J. G. and E. G. Baker in Journ. Bot. 1894, pp. 302, 345). Galium sylvestre Poll. Forest-in-Teesdale and limestone ridge near Ettersgill Beck, D., 1911. Limestone hills above Brougli, W., 1911. ■■'Valeriana Mikanii Syme. Near High Force, D., 1911. Matricaria suaveolens Buchen. Too common near Clapham, v.-c. 64. Senecio spatMilcefolitts DC. Seen in its station near Brough, Westmoreland, in 1892, in plenty, but with no signs of its attempting to flower or having flowered ! In 1911 the conditions were exactly the same, and I was interested to hear from Mr. J. Backhouse that he has never seen it in flower in this — its only inland locality in Britain — where his father discovered it. Some rosettes were brought back to Keigate in 1911 and planted in chalky soil ; these flowered sparingly in 1912, but luxuriantly and well in 1913, some of the stems bearing twenty-four heads! Mr. J. G. Baker has an interesting note upon the plant in this Journal for 1885, p. 8, where he remarks that the dividing line between S. spathulcefolius DC. and S. campestris DC. is very hard to define; in this opinion I concur. He goes on to say: " Accord- ing to Grenier & Godron, there is no campiestris at all in France, but all the French plant is spathulcefolius. But cross the Channel and immediately on the chalk downs of Sussex and the Isle of Wight campestris begins, and there is no spatkulcBfoliiis till the whole breadth of England and Wales is ci-ossed. Speaking from a geographical point of view, one would not care to believe in NOTES UPON TEESDALE PLANTS 139 such an anomaly as this, unless there were very decided evidence in its favour." As regards the first point, I think S. spatliulcBfolius (at least as far as the Holyhead and Westmoreland plants are concerned) may be best distinguished by being a larger plant in all respects (though I admit size alone is a poor character), having a more arachnoid appearance {campestris is sometimes nearly glabrous), shorter pappus, and especially by its larger number of heads and the shape of its leaves. These have on the stem peculiar broad- based petioles, whilst the root-leaves are often long-petioled (longer than the blade), with their base much more truncate than in cainjjestris. As to the second point, it may be noted that recent French botanists (Coste, Eouy, &c.) admit the two plants as inhabitants of their country, although apparently cavipestris is decidedly the scarcer. The two plants would bear further investigation, and it is worth noting that Hooker (Stud. Fl. 1884, p. 220) considers our larger plant to be S. campestris DC. var. maritima Syme (S. spathulafolius Bab. non DC). Syme (Eng. Bot. ed. 3, v. p. 90, 1866) divided S. campestris as follows : — " a. genuina. Kadical leaves entire or slightly toothed. Stem 3 inches to 1 foot high. fi. maritima. Eadical leaves generally with numerous broad teeth. Stem 1 foot high. Anthodes more numerous and larger than in var. a." But he goes on to say, "Of var. (i I have seen no specimens." Eosettes gathered in 1911, which flowered in 1913, promise to flower again this season, although some have died; Mr. Williams (Prod. Fl. Brit. i. 1901, 40) says "biennis"; most Continental floras give it a perennial habit. It may really be a biennial, whilst individuals in exceptional circumstances remain on and flourish for two or three more seasons. Owing to an error as regards county in Babington's ac- count of this plant in Journ. Bot. 1882, p. 35, Yorkshire has been credited with possessing the locality instead of Westmoreland, and this has not, I believe, been corrected until the present note. Hieracium anglicum Fr. and var. hrigantum F. J. H. Lime- stone hills above Brough, W., 1911. The latter plant was ob- served in 1892, and reported under another name in Journ. Bot. 1893, p. 219, which thus needs correcting. '■■H. lasioplmjllum Koch. Falcon Glints, D., 1892. Determined by A. Ley. H. stenolepis Lindeb. var. suh-hritannicum Ley. Limestone scars E. of Ingleborough Cave, v.-c. 64. H. sylvaticum Gouan var. ■•'tricolor W. E. L. Limestone ridges above Brough, W., 1892. Taraxacum erythrospermum Andrz. Limestone near Etters- gill Beck, D., 1911. T. p)ahistre DC. var. ■■'•runcinato-hastatttvi Lamotte (E. & F.). Above High Force Hotel towards Ettersgill Common, D., 1911 {fide J. W. White & G. Bucknall). M 2 140 THE JOURNAL OF BOTANY Camixmula latifolia L. Field-sides N. E. of Brough, W., 1911. Clapham, v.-e. 64. Gentiana campestris L. By the Tees between Winch and Shepherd's Bridges, v.-e. 65, 1911. Symphytum tuberosum L. Near High Force, D., 1911. " Kecorded from near Durham on the authority of E. Eobson, but has not been seen recently." Baker & Tate, Fl. Northumb. and Durham, p. 230, 1868. Myosotis alpestris Schm. Seen in three localities in 1892, two in Westmoreland, one in v.-c. 65 : no doubt these are the stations recorded by James Backhouse in Nat. 1884, p. 12. It seems most scarce in its Yorkshire home, where, in 1911, only two or three blossoms were seen. It is evidently considerably affected by drought, as one of the Westmoreland localities, literally blue with thousands of flowers in 1892, was not nearly such a striking sight in the hot summer of 1911. ■■'Eu2Jhrasia Kerneri Wettst. Near scars E. of Ingleborough Cave, v.-c. 64. *BhinantJnis stenophyllus Schur. Near Eibblehead, v.-c. 64. Melampyrum pratense L. var. ■'ericetorum Oliver. Stank Wood, near Appleby, W., 1911. Hills near Farther Eome, Giggleswick, v.-c. 64. Although the habit, shape of leaves, &c., seemed to bring the plants from both these localities under the montanum of Johnston, the toothed bracts seemed to oppose such a determination. This conclusion brought about a closer examination of more material and a comparison with the original descriptions of Johnston's M. montamim in his Flora of Berwick- upon-Tiveed, 136 (1829), and D. Oliwex'^ M.pratense var. ericetorum in Phytol. iv. 678 (1852). A further note by the latter writer {op. cit. 1078, 1853) seemed very much to the point, for there he argues that Johnston did not examine a full enough series to justify his specific characters of montanum, and comes to the conclusion that this plant is really but a montane form of ericetorum. This would appear to be actually the case, and thus the plants could be, I consider, more naturally arranged as M. pratense L. var. ericetorum D. Oliver, to represent the more widely spread (I believe) variety, with a "forma montanum (Johnst.) " (under ericetorum) for those who wished to differentiate the smaller-flowered, more delicate, usually montane plant with entire bracts. Utricularia minor L. Cocket Moss, near Giggleswick, not flowering ; v.-c. 64. ■■'Mentha rotundifolia Huds. Swindale Beck, Brough, W., 1911. Mr. Arthur Bennett tells me that the late Mr. Martindale found this in the county in 1907. Lamium hyhridum Vill. Koadside between Clapham and Station, v.-c. 64. Rumex domestictis Hartm, Between Eibblehead and Selside, v.-c, 64. Cephalanthera ensifolia Eich. Wooded slope, Swindale Beck, Brough, W., 1892. NOTES UPON TEESDALE PLANTS 141 Hahenaria chloroleuca Eidley. An interesting form of this occurs near High Force, with spike and individual blossoms of the size of bifolia, but the structure of the flowers seems entirely that of chloroleuca. Allium oleraceum L. var. complanatum (Bor.). Limestone W. of Selside, v.-c. 64. Scirjms setaceus L. Lane near Cocket Moss, near Giggleswick, v.-c. 64. Carex carta Good. yqx. fallax Asch. & Graeb. Mickle Fell, at about 2400 ft., v.-c. 65, 1911. C. ornithopoda Willd. Limestone hills above Brough, W., 1892. C.fulva Host. East side of Ingleborough, v.-c. 64. C. flava L. By the Tees near Winch Bridge, D., 1911.— Var. UindocariM (Tausch.). Near Malham Tarn, v.-c. 64. C. CEderi Retz. var. adocarpa And. By the Tees near Winch Bridge, D., 1911. Avena lyratensis L. var. longifolia (Parn.). Near Winch Bridge, D., 1911. ■■'Glyceria declinata Breb. Near the Ettersgill Beck, High Force, D., 1911. Lane near Cocket Moss, near Giggleswick, v.-c. 64. Woodsia ihensis Br. Seen in 1892, although as far back as 1868 it was reported as " now nearly or quite extinct." Cystoptcris alpina Desv. Seen on both visits, but in 1911 it was much cropped by sheep and no perfect fronds were noted. '■■'Polysticum Lonchitis Roth. In 1892 one plant was seen on the Yorkshire side of Teesdale ; in 1911 two or three examples in Durham territory. In Baker & Tate's Fl. of Northumb. and Durham (1868), it remarks — "Now nearly or quite extinct." Seemingly not on record for v.-c. 65. P. aculeatitm Roth. var. lobatum (Presl.). Forest-in-Teesdale, D., 1911. Swindale Beck, above Brough, W., 1911. BotrycJiium Lunaria Sw. Forest-in-Teesdale, D., 1911. Equisetum sylvaticum L. Roadside near Rome, Giggleswick, abundant, v.-c. 64. E. palustre L. An interesting form of this with branches peculiarly "flattened" (as though pressed) and prostrate, occurred at a considerable height on Mickle Fell, v.-c. 65, 1911. Mr. Arthur Bennett agrees with me in thinking it must come very near var. prostratum Hoppe, but we have not seen any authentic material for comparison. E. hyemale L. Swindale Beck, Brough, W., 1892 and 1911. Chara fragilis Desv. var. ^barhata Gant. Pond near Brough on the Middleton Road, W., 1892. — Var, '''delicatula Braun. Malham Tarn, v.-c. 64. C. contraria Kuetz. Malham Tarn, v.-c. 64. C. vulgaris L. A small form with many uncoated segments occurred in a stream between Langdon and Widdy Bank Fell, D., 1892, and a form with prominent secondary cortical cells grew in a pool near Swindale Beck above Brough, W., 1911. Nitella opaca Agardh. Pool near Winch Bridge, v,-c. 65, 1892. 142 THE JOURNAL OF BOTANY NOTES ON JAMAICAN SPECIES OF CAPPARIS. By William Fawcett, B.Sc, F.L.S., & A. B. Rendle, F.R.S. (1) Gapparis cynophallophora L. Sp. PL 504 is based on the plant Capiparis, no. 2, of Hortus Cliffortiamis, 204 ; Linnaeus merely repeats the diagnosis from the earlier work. The full description given in Hortus Cliffortianus points unmistakably to the species usually known as C. jamaicensis Jacq. ; Linnaeus cites Pluk. Aim. 126, t. 172, f. 4, as a synonym, but states that the figure differs from his own plant, and omits this reference in the Species Plantarum. A dried specimen from the Hortus Cliffortianus in the National Herbarium is without doubt Capparis, no. 2, of Hortus Cliffortianus, as indicated by R. Brown's MS. note on the sheet. Further confirmation is found in the Linnean Herbarium, where there is a specimen of C. jamaicensis Jacq. named, in the handwriting of Linnaeus, C. cynophallophora. In the Systema, ed. 10, 1071 (1759) Linnaeus enlarges the diagnosis and cites references from Plumier(Ic. 73, f. 1) and Browne (Jam. t. 27, f. 1), and in the second edition of the Species Plantarum (p. 721), which has been mostly used by the older botanists as the starting-point for Linnean names, gives additional synonyms from Plukenet and Sloane, which, with the citation from Plumier, refer to another species, the one known generally as C. cynophallophora, not to C. jamaicensis Jacq. The quotation " Brown. Jam. 246, t. 27, f. 1 " {Breynia, no. 1. Fruticosa, &c.), is somewhat doubtful, but the drawing of the calyx in the plate w'ould point perhaps rather to C. jamaicensis Jacq. than to the other species. In his own copy of Browne's Natural History of Jamaica Linnteus has written Capparis cynophallophora against Breynia, no. 1., but there is no specimen from Browne in the Linnean Herbarium. The plant of Plumier, Plukenet, and Sloane is identical with Linnasus's C. fiexuosa described on the next page (Sp. PI. ed. 2, 722), and based on Morisonia fiexuosa Amoen. Acad. v. 398. It is founded on a Jamaican specimen from Patrick Browne which is named "fiexuosa" in Linnaeus's hand in the Linnean Herbarium. Under C. siliquosa L. Syst. ed. 10, 1071, are cited two refer- ences—" Brown. Jam. 246, n. 2," and "Pluk. Phyt. t. 327, f. 6." In his copy of Browne's Natural History of Jamaica, Linnjeus has written " Capparis siliquosa" against the species in question, Breynia, no. 2. The original of Plukenet's plate is in herb. Sloane, and is C. loncjifolia Sw., apparently a distinct species, but flowers are unknown. In Species Plantarum, ed. 2, 721, Linnaeus doubt- fully includes the reference to Plukenet, and adds the remark, "Simillima praecedenti " (i.e. cynophallophora). In the Linnean herbarium there is, mounted on the same sheet with a specimen of G. cynophallophora, another specimen named by Linnaeus G. siliquosa, which is only a form of G. cynophallophora L., and supports the view that G. siliquosa is conspecific with G. cynophallophora. (2) G. baducca L., which follows cynophallophora in the Species NOTES ON JAMAICAN SPECIES OF CAPPARIS 143 Plantar urn, ed. 1, is similarly based on a reference to Hortus Glijfortianus, namely, Ca2)imris, no. 3, p. 204 ; and the expression "foiiis . . . perspatia confertis" clearly indicates C.frondosa Jacq. (Enum. PL Carib. 24), the name generally borne by this plant. The Asiatic plant referred to in Hortus Cliffortiamis as ft, from which Linnaeus took the trivial name, is another species, but Linnffius incorporated this and the references under it as synonyms of G. haducca in the first edition of the Species Plan- tarum, and in the second edition added synonyms from Plumier & Browne, which again are different species. (3) Breynia indica L. Sp. PI. 503 is renamed Gapparis Breynia in the Sy sterna (ed. 10, 1759) and in the second edition of the Species Plantarum, but according to the rules of nomenclature, the original trivial name must be restored, and the species must be cited as Gapparis indica. (4) G. ferruginea L. Syst. ed. 10, 1071. Linnseus's diagnosis is probably based on Browne's specimen, which in his own herbarium is named " ferruginea " in Solander's hand. He cites Brown. Jam. t. 27, f. 2, which in the second edition of Browne's work (in which names have been added on the plates) is rightly named Ganella alba, and there is a sheet of this species from Browne in the Linnean herbarium, which Linnseus has erroneously referred to "Baducca" (c/. Sp. PI. ed. 2, 720). In Amcenitates, v. 398, Linnaeus gives a fuller description of C. ferruginea, and in Species Plantarum, ed. 2, 721, 2, omits reference to Browne, t. 27, f. 2, but cites as a synonym " Crataeva fruticosa, &c.. Brown. Jam. 247, t. 28, f. 1," which is undoubtedly the true G. ferruginea, and is thus named in Browne's second edition. In the text of his work Browne appends his tab. 27, fig. 2, to the description of "Breynia (3). Fruticosa, &c.," which is doubtless the same as " Crataeva (3). Fruticosa, &c., t. 28, f. 1," and bears the same common name, " Mustard Shrub with a willow leaf" ; it is therefore a synonym of G. ferruginea, and in his copy of Browne's work Linn»us has written the name Gapparis ferruginea against both these descriptions. The names to be adopted for the Jamaican species and the synonymy are as follow: — 1. C. CYNOPHALLOPHOEA L. Sp. PI. 504 (1753), Hort. CHff. 204 (1737), Syst. ed. 10, 1071 (1759) (excl. syn. Plum.), and Sp. PL ed. 2, 721 (1762) (excl. syn. Plum., Pluk. & Sloan.). G. siliquosa L. Syst. ed. 10, 1071 (1759) (excl. syn. Pluk.). C. jamaicensis Jacq. Enum. PL Carib. 23 (1760), Sel. Stirp. Amer. 160, t. 101, and Ed. Pict. 78, t. 150. C. torulosa Sw. Prodr. 81 (1788), and Fl. Ind. Occ. 932. G. Breynia Sw. Obs. 210 (1791) (non L. nee Jacq.). C. emarginata A. Eich. in Sagra Cub. x. 28, t. 9 (1845). Breynia 1. Fruticosa, foiiis ohlongis obtusis. Tab. 27, f. 1. Browne Hist. Jam. 246 (1756) (excl. syn.). Breynia 2. Arborescens, foiiis ovatis utrinque acuminatis, siliqua torosa longissima, Browne loc. cit. (excl. syn.). 144 THE JOURNAL OF BOTANY 2. C. iNDicA comb. nov. C. Breynia L. Syst. eel. 10, 1071 (1759) (non Sw.). C. anygdalifolia Jacq. Enum. PI. Carib. 24 (1760). G. amyadalina Lam. Bncyc. i. 608 (1785). Breynfa inclica L. Sp. PI. 503 (1753). Ceratonia affinis arbor siliquosa d'c., Sloane Cat. 153 (1696), and Hist. ii. 60. Salix arbor folliculifera, &c., Pluk. Aim. 328, t. 221, f. 1 (1696). Breynia amygdali foliis latioribus Plum. Nov. PI. Amer. Gen. 40 (1703). Breynia elceacjni foliis Plum. loc. cit. t. 16. 3. C. LONGiFOLiA Sw. Procli". 81 (1788), and Fl. Ind. Occ. 934. Salix folliculifera longissimis argenteis ct acutis foliis Pluk. Aim. 328, t. 327, f. 6 (1696). 4. C. FEKRUGiNEA L. Syst. ed. 10, 1071 (1759) (excl. ref. to Browne), Amcen. v. 398, and Sp. Pi. ed. 2, 721. C. elaagnifolia Jacq. Enum. PI. Carib. 23 (1760). C. octandra Jacq. Sel. Stirp. Amer. 160, t. 100 (1763), and Ed. Pict. t. 149. Breynia 3. Fruticosa, foliis singularibus, oblongo-ovatis,superne nitidis, siliquis minoribus teretibus cequalibus (excl. tab. 27, fig. 2), Browne Hist. Jam. 246 (1756). Crateva 3. Fruticosa ; foliis singularibus obhngis utrinqtte acutis, subtus quasi villosis ; floribus octandris, racemis comosis alaribus. Tab. 28, f. 1. Browne op. cit. 247. 5. C. BADuccA L. Sp. PI. 504 (1753). C.frondosa Jacq. Enum. PI. Carib. 24 (1760), Sel. Stirp. Amer. 162, t. 104, and Ed. Pict. t. 153. 6. C. FLExuosA L. Sp. PI. ed. 2, 722 (1762). C. cynopliallopJwra L. Syst. ed. 10, 1071 (1759) (with ref. to Plumier), and Sp. PI. ed. 2, 721 (in part) (non Sp. PI. 504). Morisonia flexuosa L. Amoen. v. 398 (1760). Acaciis affinis arbor siliquosa &c., Sloane Cat. 153 (1696), and Hist. ii. 59. Cap)paris arborescens lauri foliis fructu longissimo Plum. Cat. 7, PI. Amer. (Burm.), t. 73, f. 1, and Ic. ined. ii. 36. Cynophallophorus, &c., Pluk. Aim. 126, t. 172, f. 4. SOME HEPATIC^ FROM THE ISLE OF MAN. By J. B. Farmer, F.R.S. The object of this communication is to draw the attention of those who enjoy facihties for a study of the Hepaticae of the Isle of Man to the circumstance that, so far as I am aware, little or nothing has yet been done towards recording the species that occur in the island. The subjoined list makes no pretence to be other than a very small contribution to the subject ; the plants SOME HEPATIC^ FROM THE ISLE OF MAN 145 were all collected during a short visit in April, 1914, to the Marine Biological Station at Port Erin, followed by a two days' walk through the more hilly part of the island. It seems, however, to be worth while to publish this note, in order that others may, perhaps, be induced to extend the list, and to deal with the problems presented by the distribution of the plants within the Manx area. More extended search will unquestionably result in the recognition of many other indigenous species and genera. Conocephalum conicum (L.) Dum., common. Lunularia cruciata (L.) Dum., Port Erin. Aneura multifida (L.) Dum., rather common. — A. sinuata (Dicks.) Dum., near Port Erin. — A. ijinguis (L.) Dum., rather common. Metzgeria furcata (L.) Dum., Colby Glen; Chasms, Port Erin. Pellia einpkylla (L.) Corda, very common. Alicularia scalaris (Schrad.) Corda, many forms of this vari- able species. Eucahjx subellipticus (Lindb.) Breidl., wet rocks in Sulby Glen. Aplozia crenulata var. gracillima (Sm.) Heeg., Laxy Glen. Gymnocolea inflata (Huds.) Dum., roadsides through heather moors. Plagiochila asplenioides (L.) Dum., Fleshwick Bay, Lophocolea cuspidata Limpr., woods and banks. — L. hetero- pliylla (Schrad.) Dum., woods and copses. Ghiloscyphus polyanthus (L.) Corda, below wet rocks behind Eleshwick Bay. Cephalozia bicuspidata (L.) Dum., common. Galypogeia Trichomanis (L.) Corda, common on the moors. Dip)lophyllum albicans (L.) Dum., common. ScajJania nemorosa (L.) Dum., Sulby Glen. — S. imdulata (L.) Dum., Sulby Glen, &c. Lejeimia cavifolia (Ehr.) Lindb., rocks near Port Erin. — L, cavifolia var. planiuscula Lindb., Colby Glen. Fmllania dilatata (L.) Dum., common on tree-trunks in damp copses and woods. — F. Tamarisci (L.) Dum., amongst moss, not uncommon. Anthoceros Icevis L., Sulby Glen. It will be noticed that several genera which might have been expected to furnish species to be included in the above list are conspicuous by their absence from it. Thus Lepidozia, Lophozia, Fossomb)Wiia, Madothcca, Badula — to mention only a few of the more prominent ones — were not encountered, although it seems hardly likely that they are really unrepresented. In any event the geological and general physical character of the island, as well as its geographical position, should serve to render the study of its hepatic flora both attractive and interesting. 146 THE JOURNAL OF BOTANY ALABASTEA DIVEESA.— Part XXIV.- By Spencer le M. Moore, B.Sc, F.L.S. (Continued from p. 98.) II. Rhamphogyne, Asteroidearum gen. nov. (Plate 530b.) Capitula heterogama, disciformia, paucii3osculosa, flosculis ext. 5 int. (? . Involucrum ovoideum e phyllis paucis sub-2-seriatis membranaceis sistens. Eeceptaculum parvum, planum, nudum. Corollae parvulae, oblongae, fll. fem. 3-dentatse, fll. hermaph. 3-4-dentat8e. Antherae 3-4, apice baud appendiculatiie, basi rotundatse, oblongo-ovoidete, cite sejunctae. Ovarium superne in rostrum deflectum excurrens. Styli til. bermaph. rami breves, complanati, appendicibus papillosis sibi ipsis fere ajquilongis onusti. Acba3nia compressa, calva, longe rostrata. — Herba perennis, nana, caespitosa, ramosa. Folia alterna, imbricata, pinnatifida vel Integra. Capitula parva, ad apicem ramorum solitaria necnon sessilia. Ehamphogyne rhynchocarpa, sp. unica. Ahrotanella rhyncho- carpa Balf. fil. in Phil. Trans. E. Soc. clxviii. 352, tab. 27 a. Hab. Eodriguez Island ; Dr. J. B. Balfour. The genus Ahrotanella being essentially Antarctic, the alleged occurrence of a species in the Island of Eodriguez seemed to be a point worthy of examination. The figure cited above, very good except for a most important omission, represents a plant that might pass muster as a somewhat abnormal member of the genus to which it is referred. The style-arms, however, of the herma- phrodite florets with their papillose appendages are not those characteristic of the tribe Anthemidece. By some oversight these have been omitted from the drawing, the only style-arms shown being those of a female floret, although the style of the herma- phrodite florets is said in the description {I. c.) to be " alte bifido, ramis ciliatis." In fact, this plant undoubtedly belongs to the Asteroidecs, and should find a place in the neighbourhood of Dichrocephala. The most peculiar feature about it is the curious beaked ovary and achene. ^ III. Decadia Lour. >, The characters Loureiro f gives for this genus are succinctly as follow. A persistent 3-leaved calyx ; corolla of ten petals; stamens about 30, adnate to the base of the petals ; superior ovary with a filiform style and somewhat fleshy stigma ; fruit a 3-celled drupe. The plant he calls Decadia aluminosa, but assigns no place to it among Dicotyledones. Blume \ suppressed Decadia in favour of Dicalyx, also a genus of Loureiro's, but printed near the end of the work already cited and a good way after Decadia. This is rather curious in view of the discrepancies in the two * The types of the species described are in the National Herbarium, t Fl. Cochinch. 315 (1790). | Bijdr. 1116 (1826). DECADIA 147 descriptions drawn up by Loureiro ; moreover, as will be seen directly, Blume could not have examined the type of Dccadia aluminosa, although he claims that name as a synonym for his Dicalyx aluminosus. Wight and Arnott "■' remark about Loureiro's plant tliat it "appears to be a species of Syviplocos." Meisner,! writing shortly afterwards, is in doubt about it, and places it with a note of interrogation in TiliacecB and in Terns trcemiacece as well as in Rosacea, finally following Wight and Arnott in thinking it a Syviplocos. Endlicher j: sinks Decadia in Symplocos, but with doubt, and the same conclusion is reached by De Candolle § on the authority of a specimen of ;S'. spicata in herb. Hamilton under the name of Decadia spicata. For Lindley |1 Decadia is synonymous with Syviplocos. Bentham and Hooker ^i are more cautious, and in the absence of material for examination consider the position of Decadia uncertain ; Baillon appears to have passed it by un- noticed; Gilg '■''■' follows Blume unhesitatingly. Brand in his monograph of Syviplocacea \\ makes Dicalyx aluviinosus Bl. a synonym of his Syviplocos aluvmiosa, but in doing this he expressly excludes Blume's synonym, i.e. Decadia aluviinosa Lour., neither can I find mention of the latter anywhere in the monograph in question. Moreover, under S. aluviinosa he remarks that Dicalyx aluviinosus has been indicated as equivalent to Syviplocos spicata Eoxb., but he cannot confirm this, as of the three Blume speci- mens named Dicalyx aluviinosus in the Leyden herbarium none is S. spicata, two being types of new species — S. aluviinosa Brand and S. syrincjoides Brand, while the third is *S. ferruginea Eoxb. The position to-day is therefore that Decadia is one of those puzzles only to be solved by examination of the types upon which they have been founded. The British Museum, as is well known, shares with Lisbon and Paris the distinction of possessing types of Loureiro's collect- ing, and among those at the first-named establishment is a sheet of specimens written up Decadia aluviinosa in Loureiro's own hand. There are no open flowers on these, but buds alone although in a fairly advanced state. Dissection shows Loureiro's diagnosis to be wrong in two vital particulars. The supposed 3-leaved calyx resolves itself into the bracts (or bract and two bracteoles) beneath the flower characteristic of the subgenus Hopea of Syviplocos, while the ovary, instead of being superior, is wholly inferior and not even half-superior, as is the case with some species of the genus. Loureiro's plant is therefore without the possibility of doubt a Syviplocos. As to the species — that is a more difficult matter. While it is certainly near S. spicata there are reasons for suspecting the conspeciticity of the two. It cannot be S. aluminosa, which is * Prodr. Flor. Ind. Or. 82 (1834). t Plant. Vase. Gen. 'n. passim (1836-43). X Gen. Plant. 1411 (1836-40). § Prod. viii. 240 (1844). II Veg. King. ed. ii. 593 (1847). •[ Gen. Plant, ii. 668 (1876). ** Engler & Prantl, Pflanzenfam. iv. i. 168 (1891). tt Pflanzenreich, 6 Heft, iv. 242 (1901). 148 THE JOURNAL OP BOTANY described as having a paniculate inflorescence (the inflorescence of D. aluminosa, if branched at all, is branched only at the very bottom) ; moreover, the pedicels are said to equal or exceed the calyx in length (those of D. aluminosa are exceedingly short) and the stamens are about forty. S. fcrruginea with broader leaves has stout ferrugineous spikes (Z). aluminosa has them very slender) and larger flowers. The description of S. syringoides fits our plant well except for its young branches being alluded to as ferrugineous-tomentose, a condition impossible to affirm in the other's case, what indications of young branches there may be pointing to ordinary pubescence. On the whole, I think it likely that D. aluminosa may be S. syringoides, and as such it has been written up provisionally in the National Herbarium. ^ IV. DiCALYX COCHINCHINENSIS Lour." Brand places this at the end of his monograph t among the doubtful species with the note: " Sine dubio optima species, sed adhuc specimina non sunt nota." There is a sheet of this in the National Herbarium, and although, the flowers being in young bud, the specimens leave something to be desired, one can deter- mine the affinity of the species with but little doubt as to the validity of the conclusion reached. Of course the organs Loureiro describes as a 3-leaved outer calycine whorl are, as in the last case, three bracts. The rest of the characters, generic and specific, given in the description are in the main correct so far as the specimens enable one to tell, except that the flowers are said to be hermaphrodites and females on different plants ; so far as I have been able to study the matter, they seem to be either hermaphrodite or male, and so to answer the " polygamo-masculi " sexual character given by Brand as of occasional occurrence in the genus. Although, owing to the early state of the flowers, the androecium cannot be properly examined, there seems no reason to doubt that our plant is refer- able to § Bohua, The description is appended. Symplocos cochinchinensis, comb. nov. Eamuhs sat vah- dis ferrugineo-tomentosis celere glabris, foliis elongatis oblongo- obovatis basi in petiolum validum canaliculatum ferrugineo- tomentosum coartatis margine apicem versus serrulatis ceterum integris chartaceis supra glabris pallideque nitentibus subtus sparsim puberulis, paniculis axillaribus terminalibusve foliis bre- vioribus ferrugineo-tomentosis, floribus secus inflorescentiarum ramulos pseudospicatis, bracteis pedicellum brevissimum occlu- dentibus late ovatis obtusis vel obtusissimis extus ferrugineo- tomentosis, calycis segmentis oblongis obtusis extus sericeis, corollae segmentis oblongis obtusis extus sericeis, staminibus circiter 80 ?, ovario semi-infero 3 ?-loculo, bacca ampulliformi glabra in sicco rugosa brunnea. — Dicalyx cochinchinensis Lour. Folia pleraque 15-20 cm. long., 4-5-6 cm. lat., in sicco viridia * Fl. Cochinch. 663. t Op. cit. 90. DICALYX COCHINCHINENSIS 149 vel plus minus rubiginosa ; petioli + 1 cm. long. Panicula circa 5 cm. long. Bracteae 2-4 mm. long. Calyx 2 mm. long. Bacca sicca 6 mm. long., summum 4 mm. lat. Affinity with S. floridissima Brand ; differing from it, inter alia, in foliage, the densely tomentose panicles and nearly sessile flowers. V. AsCLEPlADACEiE DUJE AfRICAN^C PR^TERVIS^. Schizoglossum Eylesii, sp. nov. Caule ultrametrali simplici e radice tuberoso-fusiformi stricto inferne nudo alibi sat crebro folioso inferne tereti eximieque striate glabrescente superne sub- tiliter pubescente, foliis sessilibus brevissimeve petiolatis anguste linearibus acutis margine revolutis praesertim in pag. inf. costa centrali eminente subtiliter pubescentibus in sicco arrectis, um- bellis pluribus lateralibus sessilibus 5-6-fioris, bracteis parvulis linearibus extus pubescentibus scariosis diutule persistentibus, pedicellis floribus subaequilongis pubescentibus, calycis segmentis lanceolatis acutis corolla brevioribus, corollas alte partitas lobis oblongis obtusis dorso sparsim pubescentibus intusglabris, coronse phyllis aegre ex basi columnae stamineae certe brevioris oriundis scutiformibus apice rotundatis intus perspicue bicarinatis et paullulum infra apicem appendicem lanceolatam superne incurvam sibi ipsis pauUo breviorem gignentibus, antherarum alis promi- nentibus appendicibus ovatis supra stigma inflexis. Hab. Ehodesia, Mazoe, alt. 4800 ft. ; F. Eyles, 500. Eadicis pars tuberosa 4 cm. long., summum 1 cm. diam. Caulis circa 12 dm. alt., juxta basin 2-5 mm. diam. Folia inferiora + 7"5 cm. long., 1-2 mm. lat. ; petioli dum adsint 1 mm. long. ; folia juniora gradatim imminuta, sc. usque ad 15 x "5 mm. vel etiam minus, omnia in sicco grisea. Bracteae circa 1"5 mm., pedicelli summum 5 mm. long. Calycis segmenta 2-2 mm. long. Corollas lobi 3 mm. long., 1 mm. lat. Coronas phylla 3 mm. long. ; pars basalis 1-75 x 1-25 mm. ; appendix 1-25 mm. long. Columna staminea 1*5 mm. long. Antherarum alas aegre 1 mm. long., appendices '8 mm. Pollinia anguste oblonga, "6 mm. long. ; glandula -25 mm., caudiculae -2 mm. long. This has much the appearance of S. strictissitmim S. Moore, but different flowers. Its place in the genus is next *S'. fusco- purpureum Schlechter & Eendle, which differs in habit, leaf, and corona. Fockea Monroi, sp. nov. Caule verisimiliter repente primo tereti subtiliter pubescente deinde angulato glabro, foliis oblongis vel anguste oblongo-lanceolatis apice obtusissimis ipso mucro- natis basi in petiolum brevem angustatis firme membranaceis leviter scabriusculis, cymis interpetiolaribus abbreviatis pauci- floris, bracteis rainutis ovatis acutis scariosis ut cymarum axis pedicelli calycis segmenta necnon corollas facies exterior pubes- centibus, pedicellis calyci aequilongis, calycis segmentis triangu- laribus obtusis vel obtuse acutis, corollas tubo calyce breviore lobis a basi lata oblongis revolutis asstivatione tortis, corona circa 15-fida dentibus subulatis interdum bifidis acuminatis dente inter- 150 THE JOURNAL OF BOTANY medio quam laterales majori tubo ligulis 5 elongatis integris vel bifidis fere usque ad apicem tubi eidem adnatis carinasque for- mantibus parte libera ex tubo longe eminente onusto addita ligula satis elongata etsi tubo inclusa integra vel bifida carinis quibusque memoratis infra medium tubi affixa basique integra vel dentata, antherarum appendicibus oblongis quam antherae circiter ter longi- oribus, folliculis fusiformibus glabris. Hab. Khodesia, Victoria ; Monro, 828, 837. Folia plerumque 4-7 cm. long., 8-10 mm. lat., in sicco viridia; costa media supra plana subtus eminens ; petioli 2-3 mm. long., supra excavati. Pedicelli 2-2-5 mm. long. Calycis segmenta 2 mm. long. Corollae tubus 1 mm. long., lobi 15 mm. long., juxta basin 2 mm. lat., superne 1 mm. vel etiam minus. Coronae tubus segre 5 mm. long., hujus dentes 1-1-5 mm. long. ; ligulse 4 mm. long. ; ligulge inclusaB 2 mm. long. Columna staminea 1 mm. long. Antherarum alse -5 mm. long., appendices fere 2 mm. long. Pollinia pyriformia, -25 mm. long. Folliculus 12-5 cm. long., inferne fere 2 cm. superne circa 1 cm. lat. Semina 9 mm., coma 3-5 cm. long. Differs from F. Lugardi N. E. Br., inter alia, in the small leaves, the distinctly pedicelled flowers, the longer segments of the calyx and the longer corona with different toothing. No. 942, of the same collector, also from Victoria, with closer and shorter branching, smaller leaves (2-3-5 cm. x 4-6 mm.), and somewhat reduced flowers, is a plant apparently conspecific with the above, but growing under different conditions. ' VI. CoNYZA DECURRENS Linn. Under Monenteles Pterocaulon A. P. de Candolle " remarks : "An forte Gomjza decurrens Lin. sp. 1206 eadem aut affinis ? " This in spite of the fact that Linnaeus gives India as the locality of his species, whereas M. Pterocaulon was founded on a Mada- gascar plant of Bojer's collecting. LinnjBus's description (/. c.) is short and almost valueless for purposes of identification ; so that but for the fortunate circumstance of there being specimens of C. decurrens in the Linnean herbarium at Burlington House, no answer to de CandoUe's query would be possible. Examination with Dr. B. D. Jackson's kind help of the Linnean material showed C. decurrens to be conspecific with an unnamed Pterocaulon in the National Herbarium collected by Bojer : this latter is without locality, the only note accompanying the specimens being to the effect that its Madagascar name is " Ari-androo-vavi." It now became necessary to ascertain whether Bojer's speci- mens just mentioned were referable to Monenteles Pterocaulon DC. I therefore forwarded to M. Casimir de Candolle a small scrap, together with a photograph of the sheet of Bojer's specimens taken by Mr. D. A. Rendle, with a request to that gentleman that he would kindly compare this material with the type of M. Ptero- Prod. V. 455, A NEW BERTYA 151 caulon in his herbarium. M. de Candolle was good enough to do this, with the result that he has no doubt about the correctness of his grandfather's surmise. Under these circumstances, M. Pterocaulon being unknown from India, one must conclude that Linnaeus was mistaken in supposing C. decurrens to be a native of that country. The plant appears to have become established in Mauritius, and in connection with that fact has received the name of Pterocaulon Bojeri Baker.''' The synonymy therefore stands as follows : — Pterocaulon decurrens, comb. nov. '' Conyza decurrens Linn. Sp. PI. ed. ii. 1206 (1763). Monenteles Pterocaulon DC. Prod. v. 455 (1836). Pterocaulon Bojeri Baker Fl. Maur. 164 (1877). Besides Bojer's material, the species is represented in the National Herbarium by the following : — North-west Madagascar, Pasandava Bay ; Ilildehrandt, 3014. Central Madagascar ; P^ev. B. Baron, 1321. North Madagascar; ibid., 6461. Description of Pi,ate 530. A. Muschleria augolensis. 1. Open inflorescence. 2. Congested ditto. 3. A floweiing capitulum, x 4. 4. A floret, x 5. 5. Anthers, x 10. 6. Ripe achene, x 16. B. Rhamplwgijne rhyncJiocarpa. Style-arms, x 16. A NEW BEETYA. By E. a. DiJMMER. While engaged on the Conifers of the Lindley Herbarium, Cambridge, my attention was drawn to a specimen included among the Podocarps of that collection, which was doubtfully referred to that genus by Lindley. This plant was collected by Eraser in the Barrens, north of Arbuthnot's Eange, in Australia, and turns out to be an undescribed species of Bertya, of the natural order Eujjhorbiacece, for which therefore the name B. neglecta is proposed. Bertya, a genus comprising approximately fourteen species, is limited to Australia and the adjacent island of Tasmania, and includes a series of shrubs, some of an ericoid aspect, of which B. gummifera merits cultural attention. The species under consideration, of which the female flowers are as yet unknown, has the superficies of B. rosmarinifolia, but is immediately distinguished from it by its more glabrescent character, the presence of minute spinules which clothe the twigs and the leaves, the relatively larger male flowers and the ten larger bracts which subtend them. Bertya neglecta, sp. nov. A bushy heath-like shrub. Cur- rent year's twigs subumbellately disposed, erect, straight and Fl. Maur. 164. 152 THE JOURNAL OF BOTANY rigid, usually unbranched, 4-12 in. long, terete or angulate towards their extremities, dull brown, spinulose and sparingly pilose, densely leafy. Leaves ascending and subimbricate, or eventually spreading, borne on broad flattened petioles scarcely exceeding ■J-^ in. long, the blade linear, obtuse or truncate at both ends, i~l in. long, averaging J^ in. in breadth, thickly coriaceous, minutely and sparingly spinulose, the upper surface with a median groove, decidedly convex on account of the revolute thickened and entire margins, the lower surface sparingly pilose and spinulose with a broadened and conspicuous midrib. Flowers solitary in the axils of the upper leaves, arranged racemosely, the inflorescences up to 2f in. long. Male flowers when unopened broadly ovoid, when mature cylindric and l-j% in. long, shortly pedicellate, sub- tended by ten imbricate persistent bracts, the lower six small oblong or triangular and thickened, the upper four thinner in texture and larger, ovate to oblong, rounded or subacute, and up to Jjj in. long ; anthers crowded and spinally arranged on a thickened ascending axis. Bertya neglecta, sp. nov. B. rosmarinifolicB Planchon affinis sed ramulis foliisque minute spinulosis glabrioribus, floribus $ majoribus, bracteis 10, difTert. A NEW ARCTOTIS. By R. A. DiJMMER. Arctotis Scullyi, sp. nov. Fruticosa, sparsim ramosa, ramis validis dein breviter subhispidulis, foliis sessilibus planis lanceo- latis vel lineari-oblongis acutis nisi acuminatis integris vel rari- dentatis lobatisve crasse coriaceis utrinque (ut scapus) arete subhispidulo-pubescentibus, scapo terminali monocephalo, in- volucri phyllis exterioribus parvis triangularibus caudato- acuminatis interioribus multo majoribus oblongis obtusis cite glabris. A shrubby plant of unrecorded height, sHghtly hispid all over and with stout angled sordid brown scabrous shortly harsh-haired current year's twigs, the internodes being i-l in. long. The leaves are ascending and often overlap, have a broad insertion but do not clasp the axis, are flat, lanceolate or linear oblong, acute or acuminate, usually broadest about or below the middle, 1^-2* in. long, i-^ in. broad, thickly leathery, and thickly clothed on both sides with a brownish or whitish short harsh pubescence, penninerved ; the margin is entire or possesses two obscure teeth or lobed about or above the middle. Scape terminal, short, 1-3 in. long, bearing one flower-head, terete, harsh-haired, nude or with a few small leaf-like bracts towards its base. Flower-head averaging 1| in. across, the lower involucral scales small, triangular and tipped with a hairy caudate acumen, the upper much larger, oblong, scarious, concave, obtuse or rounded at their apices, glabrous at maturity, up to ^^ in. long, SHORT NOTES 153 and i-i in. broad. Eay-florets averaging 1 in. long, yV~to i'^- broad, probably whitish. Arctotis Scully i falls in the section Euarctotis, characterized by the tuft of hairs which arise from the base of the achene and surrounds the latter, and is probably most closely related to A. belUdifolia, from which, however, its sessile not basally clasp- ing leaves, their scabridity on both surfaces, and their different conformation differentiate it immediately. Its discovery is due to the efforts of Mr. W. C. Scully (no. 221; ex herb. Bolus, 9615), who, during a several years' sojourn in Little Namaqualand, collected numerous plants, many of which still await description. The region alluded to is one of the richest as regards annuals in South Africa, and includes in particular an abundance of Compo- sites, many of which are of singular beauty, and would assuredly repay the trouble of introduction. SHOBT NOTES. Leucojum vernum L. and Galium Vaillantii DC. in Somer- set.— For some years Miss M. A. Hellard has known the Spring Snowfiake in a locality between Bishop's Lydeard and Williton, v.-c. 5, where she kindly showed it to me in good flower towards the end of last February. Time did not allow me to make a thorough search ; but a friend who went down to see it, a few days later, found that it extended (by and near a brook) for almost a third of a mile. It has the appearance of a native ; but I am not yet sure about its true status. The Bedstraw, previously known in this county only as a casual at Twerton, near Bath, was plentiful in potato-fields, &c., near Ashcott Station, and also occurred near Shapwick Station, v.-c. 6, last September. — Edward S. Marshall. The Early Season. — On May 1st I gathered in a wheat-iaeld on stiff clay in the parish of Norton, Herts, flowering and fruiting specimens of the Corn Gromwell (LitJiospermum arvense) ten inches high, and with well -developed fruits. Some of the white blossoms had a slight pink tinge, whereas in the mountains on the Continent they are not infrequently pale blue. Hawthorn was in blossom by April 28th on Norton Common and elsewhere in the vicinity. — H. S. Thompson. REVIEWS. Researches on the Irritability of Plants. By Tajadis Chunder BosE, M.A., D.Sc. Longmans, Green & Co. 1913. In this publication Prof. Chunder Bose has added to his previous work in this important field of physiology a faithful account of a unique series of experiments, rendered possible by his own ingenious invention of recording apparatus. Journal of Botany. — Vol. 52. [June, 1914.] n 154 THE JOURNAL OF BOTANY The author states it to have been his aim to make the plant accurately self-recording and subject to automatic stimulus, so as to eliminate as far as possible the personal factor from the experi- ments. Only those with like aims can appreciate to the full the success that Prof. Chunder Bose has attained in this direction, but the veriest tyro must be struck with the beauty and pre- cision of his automatic inventions — the oscillating recorder and the resonant recorder. In both, friction between the writing point and the writing surface is overcome by intermittent contact ; in the former by vibration of the writing surface, in the latter by vibration of the writing point, The resonant recorder was invented to provide a means of recording minute time- measurements of hundredths of seconds. This can be done very accurately by means of the intermittent recorder, providing that the oscillation period remain constant. The author has very ingeniously made use of tlie principle of resonance and ensured regularity of contact by timing electro-magnetic impulses to synchronize with the natural frequency of the recording index. Prof. Chunder Bose has also used greatly improved methods of stimulation, capable of quantitative variation, and including various electric and electro-thermal methods of excitation. With these improved methods he has been able to analyze experimentally the phenomena of stimulus and response, as seen in the sensitive plants Mimosa, Biopliytum, &c. One of the most successful results thus obtained is the demonstration of a latent period strictly comparable to that shown by animal muscle under stimulation. The ordinary fall of Mimosa leaf under stimulation is accom- panied by development of a galvanometric negativity, but it is found that, when the stimulus is applied at some distance from the responding organ, a slight positive response appears before the much more marked negative one. The author claims that his experiments show that all stimulations are dual, but that unless separated by time of travel, the stronger masks the weaker effect. Prof. Chunder Bose's main thesis is that the plant exhibits true physiological phenomena of excitation, conduction and con- traction, and that there is the most striking parallelism between muscle action and plant movement. Thus the plant shows the additive effect of repetition of minute stimuli ; increase of response with increase of stimulus, and, most striking of all, a complete reversal of direction of movement after over-stimulation. The latter fatigue phenomenon strictly parallel with relaxation of muscle on over-stimulation. Our author, believing Pfeffer's results to have been incon- clusive, has paid particular attention to the question of the trans- mission of stimulus, and claims, and we think rightly, that the present more extensive experiments prove the co-operation of the living protoplasm in the process ; in short, that it is a physio- logical and not merely a physical conduction. AN ACCOUNT OP THE MORISONIAN HERBARIUM 155 One cannot but wish that circumstances had allowed of the author discussing the results from the point of view of general plant physiology, as there are many interesting aspects which would bear comparison with other physiological processes, as, for instance, the accumulative effect of stimulation, also seen in geotropic movements ; the apparent presence of an optimum, &c. The author has noted a depression of excitability on rainy days, which he attributes to absorption of water by the pulvinus. One is not quite convinced that there may not be another inter- pretation of the results following upon application of water to the pulvinus, and still more so in connection with retardation of excitability on rainy days. Prof. Chunder Bose is much to be congratulated on furnishing such valuable additional data — data which must be taken into account in all future considerations of this difficult subject. They should prove of interest and value to animal as well as plant physiologists, and we can but hope that he will still further add to the debt we owe him by investigating the nature of the diffe- rential excitability to which he attributes the movement of the pulvinus. -^ -^ rp An Account of the Morisonian Herbarium in the possession of the University of Oxford, together ivith Biographical and Critical Sketches of Morison and the tioo Bobarts and their Works and the Early History of the Physic Garden (1619-1720). By S. H. Vines, M.A., F.E.S., Sherardian Professor of Botany in the University, and G. Claridge Druce, Hon. M.A., Curator of the Fielding Herbarium. 8vo, cloth, pp. Ixviii, 350, with portraits. Oxford : The Clarendon Press. Price 15s. net. This is a companion volume to the Account of the Dillenian Herbaria which was issued by the same Press in 1907, and was noticed in this Journal for that year (p. 282), and is yet another tribute to Mr. Druce's well-known energy ; for though Professor Vines, who in the earlier volume appeared as editor, is here placed as joint author, it may safely be assumed that the bulk of the undertaking has fallen to Mr. Druce's share. The book consists of two parts, the first containing the history of the Oxford Garden, with full biographies of the two Bobarts (1599 ?-1680 ; 1611-1719), whose names are so intimately asso- ciated with its foundation and history, and of Robert Morison (1620-1683), the first Professor in the University. These bio- graphies, each of which is accompanied by a portrait, are exceed- ingly well done; the history of the various publications of Morison and the younger Bobart is worked out in so much detail and with so great care that it can hardly be expected that future commentators will be able to supplement it. One small detail may, however, be added : the original figures in Morison's Plan- tariwi Historia were, according to Stokes (With. Arr. ed. 2, 1), " chiefly by Bobart," the editor of the book. It may be well to supply the reference to the " collection of British plants, made 156 THE JOURNAL OF BOTANY chiefly from Oxen, made by Plot and named by Bobart," which are indicated as " in the British Museum (Nat. Hist.) " : these are in vol. 113 of the Sloane Herbarium ; other plants connected with Plot are in H. S. 168. The identification of the plants of the Morisonian Herbarium is preceded by an account of the Herbarium itself and a list of the collectors whose plants it contains. These largely correspond with the contributors to the Sloane Herbarium, formed somewhat later, an account of which it is known will be published by the Trustees of the British Museum, though not as immediately as was suggested by a note inserted in this Journal (1913, 316) during the absence of the Editor. The Herbarium consists of about 6,500 specimens, including many not named in the Historia but identified here ; we can well believe that their identification has been " a long and laborious task," even when alleviated by the expert assistance which the authors acknowledge. Whether it was worth all the trouble that has been expended upon it is, of course, a matter of opinion ; the authors point out that as the col- lection was not formed by Morison but was " organized by the younger Bobart," " it might be consequently inferred that none of the specimens illustrating [Morison's] volume can be regarded as ' type-specimens ' ; but there can be no doubt that Bobart was quite familiar with Morison's species, sufficiently so as to render the specimens authoritative for reference." We ourselves do not think the identification of first importance, inasmuch as a cursory inspection of the detailed list shows that in many instances more than one plant is represented under the same name: thus " C. Madraspatensis panicula sparsa " is represented by two if not three species (p. 116), and a note on the following page on a specimen identified as " Juncoides niveum (L.) "'•' shows " a double error " of a complicated kind. There is, we cannot but think, considerable danger of basing too much upon the specimens and their identification : thus, without necessaril}- endorsing Dr. Eobin- son's conclusions as to the nomenclature of Oxalis corniculata — his paper is in Journ. Bot. 1906, not " 1907" — we do not consider that they are disposed of, as the authors (p. 13) seem to think, by the correction of one of the synonyms quoted by Linnaeus. A more thorough examination of the book would doubtless afi'ord further matter for remark, but the publishers by issuing the volume uncut have not rendered its contents readily consultable. There is an excellent index of plant-names, but none of the collectors cited nor of the English localities or counties in which British specimens were collected ; both of these would have been useful to those who consult the book, wdiich, whatever opinion may be formed as to its positive utility, must certainly take rank as a monument of industry. * This appears to be a new combination, as do " Juncoides sylvaticum (Huds.) " (p. 113), " Centauriiim spicatuni (Pers.) " (p. 57) and others similarly indicated ; these, we presume, will have to be cited as of " Vines & Druce." PLANT PHYSIOLOGY 157 Plant Physiology. Vorlesungen ilher Pjianzen-physiolotjic. By Dr. Ludwig Jost. Third edition. Pp. 760 + 11 plates+ 188 figs. Jena: Gustav Fischer. 1913. Price 16 marks. Plant Plujsioloijij. By Dr. Ludwig Jost. Supplement to the English Translation by E. J. Harvey Gibson. Pp. 168 + 7 figs. Oxford: The Clarendon Press. 1913. Price 2s. &d. net. The third edition of Prof. Jost's well-known work on Plant Physiology is very welcome to botanists. Immediately on its first appearance in 1904 it was recognised as the best general survey of the subject within the moderate compass of a single volume. Of course, the work did not compete at the time with the fuller treatise of Pfeffer, though it was certainly far more readable. However, the last part of Pfeffer's work was pub- lished in 1904, while Jost's book has been kept steadily up to date and so is now indispensable as a presentation of the physio- logical work of recent years. The new edition is well up to the standard of the two earlier ones, for the very voluminous litera- ture of the last few years has been, in the main, admirably dealt with. One finds an adequate account, within the space available, of most important recent investigations, such as that of Willstatter on the chemistry of chlorophyll, the new work on the nature of the respiratory process by Palladin and others, investigations on the permeability of protoplasm, &c. The illuminating results of various workers in the United States on the wilting coefficient are, however, not adequately treated by the statement that two workers have confirmed in the case of wheat Sachs's results with tobacco. Sachs was certainly a pioneer in this work, but he touched only the fringe of the subject, and his results should certainly give place to those of recent workers. x\gain, in discussing the question of the control of transpiration by stomatal movement, fuller reference should have been made to the recent work of Lloyd and Francis Darwin. Such small irregularities of treat- ment are inevitable in a work of this scope ; the wonder is that the general level is so high, since the labour of compilation must have been very heavy. The bibliographical references, instead of following each chapter as in previous additions, are now distributed as footnotes in the text ; this is a great improve- ment. The second book is a supplement to the English translation of Jost's work. That it should come for review at the same time as the third German edition does little credit to the Clarendon Press. The English edition was a translation of the first German edition, and appeared in 1907, in the same year as the second German edition. There were a number of mistakes in the English edition, and it was, of course, partly out of date very shortly after its appearance. It has taken six years to rectify the mistakes and bring the English edition up to the level of the second German 158 THE JOURNAL OF BOTANY edition by means of this supplement which incorporates the alterations in that edition. As a result of this long delay, which seems inexcusable, the supplement appears at the same time as a new German edition, and the English rendering again falls behind. Apart from its being already out of date, the supplement is very troublesome to use, as it has to be compared page by page with the English edition. Nothing less than an early translation of the third German edition will now be satisfactory to botanists, nor, we may add, consonant with the dignity and reputation of the Oxford Press. ,, -tt r, V. xi. a. BOOK-NOTES, NEWS, dc The "Report for 1913" of "The Botanical Exchange Club and Society of the British Isles" becomes yearly less and less obviously connected with the Club and more and more a medium for the expression of the views of Mr. Druce, the Secretary, upon various botanical matters and above all for the publication of the " comb.-nov." of which he is so expert — we had almost said so unscrupulous — a manufacturer : the stronger expression was sug- gested by finding (p. 314) three new names suggested for the same plant ! We have more than once expressed an opinion, which we know is widely if not universally shared, as to the unfortunate obsession which can regard work of this kind as in any way tending to the advance of science, and it seems useless to protest further against Mr. Druce's action. Suffice it to say that in the present "Report" he seems to have surpassed himself, and the proposed recognition of John Hill's accidental binominals has afforded him a new opportunity for the display of his powers in this direction. The " Report " has not reached us for notice, so we are excused from saying anything more about it. It is note- worthy that Mr. Druce says that his opinion "in no way assumes to carry with it the authority of the Club." From a printed circular which, though headed " private and confidential," is sent out with the " Report," we learn that Mr. Druce is " anxious to raise a sum of £200 in order to be able to publish a history of the London Botanical Society, to reprint a few of the earlier Reports of the Club, to publish a general Index to the whole of the Reports, to put the Society in a more satis- factory financial position." As to the first object, assuming that by " the London Botanical Society " the Botanical Society of London is intended, we are glad that Mr. Druce proposes to act on the suggestion made in this Journal for 1911, p. 352 ; " no one," as we then remarked, "could do it better." The others seem to us of more doubtful utility : the last we do not understand — "the Society " cannot be that previously mentioned in the same sentence: does it refer to the "Society of the British Isles," which sprang into existence as a nomcn nudum on the titlepage of the Exchange Club Report for 1910 (see Journ. Bot. 1911, 325), BOOK-NOTES, NEWS, ETC. 159 and has, so far as we know, never received the diagnostic character necessary to recognition ? The Annals of the Bolus Herbarium is a new periodical, of which two parts (price 5s. net each) will appear each year. It is a small quarto, edited by Dr. Pearson, the Professor of Botany of the South African College, Capetown, in connection with which it is issued : the present number contains forty-four well-printed pages and six plates executed by some not very pleasing pro- cess : the figures themselves however are informative and useful. A portrait of Dr. Bolus appears as frontispiece : the papers are on the Flora of the Great Karasberg, with an introduction by Dr. Pearson and a list of the plants collected by F. and L. Bolus and M. W. Glover, to which the plates relate ; a description of a new genus of Iridcce (Pillansia) by F. Bolus, which is not figured ; a Key to the Spermaphyta of the Cape Peninsula ; and a review of Dr. Moss's Vegetation of the Peak District, which doubtless " contains much that should be of interest to students of the vegetation of S. Africa," but seems somewhat out of place here. At the meeting of the Linnean Society on March 5th, a paper was read by Mr. C. F. M. Swynnerton, entitled " Short Cuts by Birds to Nectaries," illustrated by lantern-slides from photo- graphs and drawings made by the author during his travels in Africa. He stated that birds were watched visiting flowers, and flowers were examined for indirect evidence. Not only sunbirds (which indeed are often great evaders of pollen), but many other birds as well, visited certain flowers freely for their honey, and were probably of use to them for cross-fertilization. Certain birds, and some individuals more than others, apparently disliked being besprinkled with pollen, and tended always to enter flowers by breaches made by themselves or their predecessors. Other birds tried, contrariwise, to enter the flowers by their natural openings and so to be of use to them for cross-fertilization, excepting in the case of individual flowers that happened, through inconvenience in their own or the bird's position, &c., to ofl'er some difficulty. If these were insufiiciently protected as well, they were often either pierced or the openings already made in them by the more indiscriminating birds were utilized. Insects also tended to utilize the breaches made by birds, and so probably in large part failed to counteract the latter's discriminative influence. In most cases the eliminative effect, if any, of the damage was not traced. In two instances it was (for individuals) immediate and clear, flowers of a certain type being bodily removed. At the meeting of the same Society on April 2nd, Mr. E. Allen Kolfe, A.L.S., exhibited a series of coloured drawings of five hybrid Ophryses, raised by M. Fernand Denis, Balaruc-les- Bains, France, from Ophrys tenthredinifera, Willd., crossed with the pollen of 0. aranifera Huds. ; together with the two parents. This is believed to be the first hybrid Ophrys raised artificially, IGO THE JOURNAL OF BOTANY and it proves the origin of a natural hybrid that has been recorded from three locahties in Italy, and is known under the names of 0. Grampinii Cortesi, and 0. etrusca Asch. & Grabn. The hybrids varied somewhat between themselves, but all showed an unmistakable combination of the characters of the two parents, particularly in the colour and markings of the lip, and in the peculiar combination of rose and green in the sepals and petals. M. Denis has a batch of some forty seedlings in flower or bud. At least eighteen natural hybrid Ophryses have been recorded, and Mr. Eolfe believed there were others. He would be greatly obliged to anyone who would send him examples at the Kew Herbarium, as he is studying them. i\.T the meeting of the same Society on May 7th, Mr. H. N. Eidley gave an account of "The Botany of the Utakwa Expedi- tion, Dutch New Guinea," which had been worked up by various botanists. He stated that the extensive collection of plants made by Mr. C. B. Kloss during Mr. Wollaston's expedition to Mount Carstensz, Dutch New Guinea, in 1912-13, is the most important collection of New Guinea plants brought to this country. In spite of the large collections made by Dutch and German collec- tors, there are upwards of five hundred new species and eight new genera in the collection, many of great interest. The plants were collected at various heights from sea-level to an altitude of about 13,000 feet, where vegetation ceased. The areas explored may be divided into four botanical regions: — (1) The Coastal region, where the flora was largely of Malayan affinity. (2) The foot- hills, ranging from 500 to 3000 feet elevation, an area of dense forest, the flora still typically Malayan but containing a distinct Australian element. (3) The Frontal mountain belt from 3000 to 8000 feet elevation, the begonia and balsam region. Here culti- vation ceased. Palms disappear, and the first of the Palsearctic forms are met with, such as Viola, Banunculus, Hypericum, and Galium. (4) The main mountain range. Here the big forest trees disappear, and herbaceous plants show a marked increase. Casuarinas, Pandani, and Violets form a conspicuous part of the flora. The highest tree is Podocarpus papuanus, sp. nov. This attains an altitude of 10,500 feet. Above 11,000 feet the rocks became too steep for most plants, the only plants being rhodo- dendron bushes, a daisy {Myriactis), some grasses and mosses. The flora of this upper region from 8000 feet upwards comprised many Palaearctic forms. Geranium, Thalictrum, Astilbe, Euphrasia, Potentilla, Gentiana, &c., with the Australian types, Pterostylis and Corysanthes. We regret to announce the recent deaths of Mr. William West, Mr. J. A. Martindale, and the Eev. E. N. Bloomfield, of whom notices will appear in due course. The address of the Eev. H. J. Eiddelsdell is now Wigginton Eectory, Banbury. Lafayette. Manchester, WILLIAM WEST, F.L.S. 161 IN MEMORY OF WILLIAM WEST. (1848-1914.) (with portrait.) By W. Denison Roebuck, F.L.S. By the death of WiUiam West, which took place at Bradford on May 14th after a brief illness, this Journal has been deprived of one of its most valued contributors. His first paper, " Bryo- logical Notes," appeared in 1881, and from that time until 1912, when he published a long and interesting paper on the Flora of Shetland, "with some ecological observations," hardly a volume has appeared without a contribution from his pen. As these contributions — a small portion of his literary output — show, he was a man with an extraordinarily wide and varied range of infor- mation. He had a competent knowledge of all branches of field botany, and his attainments in plant physiology and morphology showed that, had he been specially interested in those branches of study, he would have made his mark as an original investigator. But it was as a student of the freshwater algae, and especially of the Desmids, that he obtained his world-wide reputation. In this department he was one of the foremost men of his time, and the numerous papers and memoirs contributed to various journals and to the Transactions and Proceedings of learned societies testify to his unflagging energy and zeal in the pursuit of his favourite study. As a systematist he has been for many years recognized as an authority on the freshwater algae, and he has also made valuable contributions, in numerous memoirs, to our knowledge of their distribution and biological relationships. WilUam West was a native of Leeds, a city which has pro- duced not a few naturalists of distinction, and was born February 22nd, 1848, on the edge of Woodhouse Moor. He studied for the pharmaceutical profession, eventually quaUfying and being registered on November 16th, 1870, and removing to Bradford in 1872, set up in business there. He was married in 1874 to Hannah Wainwright, also a native of Woodhouse Moor, Leeds, who died in 1904, leaving two sons and a daughter, all of whom inherited their father's ability, the sons passing through Cambridge University with high distinction, and both of them taking up botanical work. The elder, William, died in India in 1901 (see Journ. Bot. 1901, 353) ; the younger, George, is now Professor of Botany at the University of Birmingham. In 1886 William West took up science teaching as a profession, and was appointed Lecturer in Botany, and afterwards also in Biology and Pharmacology at the Technical College, Bradford. He was remarkably able and successful as a teacher, gaining the respect and affection of his students to an extraordinary degree. The Yorkshire Naturalists' Union was established on its present lines in 1877, and West was one of the band of able naturalists who were instrumental in making it the powerful and Journal of Botany. — Vol. 52. [July, 1914.] o 162 THE JOURNAL OF BOTANY successful instrument of local scientific research which it has been ever since. He became Secretary of the Botanical Section in succession to Dr. H. Franklin Parsons, but his professional duties prevented his taking much active part in its work after these first few years. He was elected President for 1899, a signi- ficant mark of the appreciation of the esteem in which he was held by his fellow Yorkshiremen. He became a Fellow of the Linnean Society on March 17th, 1887; he was also a member and frequent attender at the meetings of the British Association, and Secretary of its Botanical Section in 1900 at Bradford. In the earlier years (1878-1887) of West's scientific work he was an all-round botanist with a wide and accurate knowledge of all the groups both of flowering and flowerless plants, being gifted with a powerful and retentive memory and remarkable powers of observation. He published numerous notes about this time, dealing wath such subjects as Mosses (1878), the Autumn Flora of Whernside (1879), a February stroll near Baildon (1881), the principal plants of Malham (1883), the plants of the Bradford district (1886), Buckinghamshire Lichens (1880), the Eoses of Towton Battlefield (1879), &c. He contributed a considerable amount of material to Lees's Flora of West Yorkshire (1888), after which he began to concentrate his energies on the freshwater algae, the Desmicliacece especially. His son George was now co-operating in these studies, and the practical self-training of the father and the parental and academic training of the son, based upon a combination of practical field-work and an apprecia- tion of specific and varietal differentiation with a capacity for broad and sound generalization, began to yield fruit in no small degree. Theirs was no mere local study, the whole world was now their sphere of investigation, and the command of the com- plete literature of their subject and of innumerable gatherings from almost all parts of the globe, with the willing co-operation of European and American workers, enabled the two Wests to establish themselves among the foremost students of their subject. William West's remarkable knowledge of cryptogamic plants of all kinds and of their conditions of growth made him a unique personality in Britain, probably in Europe. He was an ecologist long before the term itself was coined, always fully conscious of the importance of the common and dominant forms. The algo- logical investigations which were now his main line of research were most systematically and diligently carried on. Holidays were utilized to the full for visiting all parts of the British Islands, especially the outlying montane regions of Scotland and Ireland, North Wales, and the English Lakes. The work began near home, and their native county of Yoi'k was worked, a list published for each of its Eidings, and finally in 1900-1901 a complete alga- flora of the county. Then came papers dealing with North Wales (1890), the English Lake district (1892), the West of Ireland (1892), Scotland (1893), the South of England (1897), the North of Ireland (1902), the Orkneys and Shetlands (1905), and the Clare Island Survey (1912), many of which were published in this IN MEMORY OP WILLIAM WEST 163 Journal. European countries were left to Continental workers, except for papers dealing with Denmark (1891) and Portugal (1892). But material was sent to them from many parts of the world, and this formed the basis of memoirs published for the American States of Maine (1888, 1891) and Massachusetts (1889), and for the West Indies (1894, 1899). For the Old World were published papers on Singapore (1897), Koh Chang (1901), Ceylon (1902), Burma and other parts of India (1907), and Kinabalu and North Borneo (1914) ; meanwhile another able Leeds algologist, Mr. W. Barwell Turner, had monographed the Desmids of India. The Wests dealt with Madagascar in 1895, Central Africa in 1896, and Welwitsch's African collections in 1897, and (in 1911) the freshwater algae collected by the Shackleton Antarctic Expedition. Besides these more serious undertakings, numerous notes were published in various journals, as well as articles of more general scope and import, including memoirs on the Conjugation of the Zygnemaceae (1891), and Observations on the Conjugatse (1898). Speaking generally, the earlier papers (to 1893 or so) appeared as by William West alone, the later ones by himself and his son jointly ; but the co-operation in the work had extended over the whole series, and of later years the algological work fell to the son, while the father devoted time to the study of the ecology of the bryophytes and lichens. Finally came the pubHcation of their culminating work by the Eay Society — the Monograjyh of the British DesmidiacecB ; of this four volumes have appeared (1904, 1905, 1908, 1911), while two remain to be completed by the surviving author. These algological investigations did not, however, exhaust the potentialities of the subject, and led up to another line of in- vestigation, that of the phytoplankton of lakes and rivers. In this the two Wests were the pioneer British workers, and they took up the task in characteristically full and systematic fashion. Aided by grants from the Government Grant Fund and from the Eoyal Irish Academy, the detailed field work was begun about 1900, and Western and Southern Scotland, the lakes of England and iSforth Wales, those of Western and South-western Ireland, as well as Lough Neagh, Malham Tarn, and the rivers Ouse, Lochay, and Bann, were visited during the vacation seasons of several years. The results of these plankton researches proved to be of high importance, and were summarized in the Proceedings of the Boyal Society for 1909. From a bio- logical point of view the British Lakes are of great interest, the researches of the two Wests showing that the lake- plankton of extreme Western Europe, and particularly of the British Islands, differs completely from that of Central Europe, being characterized by the presence and dominance of Desmids. Their observations showed that Desmid -plankton occurred only in rich Desmid-areas, and that these rich areas were directly correlated with montane areas, with heavy and persistent rainfall, and, most important of all, with the presence of the oldest rocks, Archaean and the older palaeozoic rock-formations ; and their 0 2 164 THE JOURNAL OF BOTANY success in working out this new line of research has produced significant results which were a revelation and a surprise to Continental observers. The last subject which occupied West's attention was the ecology of cryptogams, on which a paper — the first of a projected series — was read in abstract at the Linnean Society at the meet- ing on June 18, a month after his death : of this an abstract will be found on p. 191. The personality of William West endeared him to all with whom he came into contact. He was a man of warm enthusiasms, with a singular charm of manner and a quiet vein of genial humour, and those who, with the present writer, have been on most intimate terms with him for nearly forty years, can best appreciate what manner of man he was, and feel the greatness of the loss which they have sustained by his succumbing to heart failure supervening upon an attack of his old enemy asthma. He was followed to his resting place in Schotemoor Cemetery, Bradford, by a great number of his old students and his old friends. The accompanying portrait is reproduced from a photograph taken by Messrs. Lafayette, of Manchester. NOTES ON SOME PLANTS OF MID-PERTH. By Rev. E. S. Marshall, M.A., F.L.S. The greater part of last July was spent at Fortingal, v.-c. 88, where there is an excellent hotel ; Messrs. W. A. Shoolbred and C. E. Salmon joined our party for most of the time, and we had the pleasure of again meeting Mr. D. A. Haggart, who accom- panied us on several occasions. At the beginning of September Dr. C. E. Moss and I had three days' collecting in the same district, which is quite rich, as Messrs. Lintons' papers of over twenty years ago indicated. This vice-county has, I suppose, been better worked than any other in the Highlands, so that novelties must be few and far between ; such supposed additions are starred. The Lyon Valley hereabouts is rich in Thalictrum. I doubt whether restricted T. minus L. {T. collinum Wallr.) occurs there ; but a plant closely allied to T. majus auct. angl. is frequent, as well as the one referred by the Lintons to T, Kochii Fr., of which it has the ovoid fruit. ■■' Caltha raclicans Forster. One luxuriant specimen occurred by a streamlet on the south side of Ben Lawers, at about 2000 ft. ; usually it is a low-ground plant. Erophila inflata Hook. fil. ascends to fully 2600 ft. on Ben Lawers ; the pods are shorter and often less inflated than in my gatherings from Glen Shee, E. Perth, and between Altnaharra and Tongue, W. Sutherland, perhaps owing to the greater altitude. Cochlearia micacea E. S. Marshall, Abundant on the north NOTES ON SOME PLANTS OF MID-PERTH 165 side of Meall Garbh ; very local on the north side of Creag Mohr. It occasionally descends to about 2600 ft., and on Ben Lawers reaches 3950 ft.; but the main range is between 3000 and 3500 ft. The long-podded form is scarce on these mountains ; a very few pink-flowered plants were observed. In cultivation I find it to be a rather short-lived perennial ; but I suspect that it is usually biennial, in the wild state. The dense cushions of early bloom are very striking and pretty. Arenaria rubella Hook. A specimen from the west side of Ben Lawers measures five inches across. A. leptoclados Guss. var. viscichda Eouy & Fouc. Dr. Moss detected this on a roadside wall near Coshieville. Hypericum maculatum Crantz [duhium Leers). What appears to be a form of this, very different from our ordinary EngHsh one, being of low growth, with much paler flowers, was found in good quantity by the roadside between Fearnan and Lawers, and sparingly near Bridge of Lyon, Fortingal. Vicia sylvat'ica L. Very fine by the Allt Odhar, Fortingal, and by the x\llt Coire Pheiginn, at 700 ft. ; not observed elsewhere. Agrimonia Eupatoria L. We collected (for A. odorata Mill.) flowering examples of a luxuriant Agrimony, fairly plentiful among bushes and rocks on the cliffs above Fortingal, from 500 to 1100 ft., with large, glandular leaves ; but Mr. Arthur Bennett suggests A. Eupatoria var. '•'sepnim Brebisson, Fl. Norm. 110 (1869), which may be the same as var. umbrosa Coss. & Germ., Fl. Paris. 182 (1845). Galium erectum Huds. One fine plant on limestone near Garth Castle; Mr. Haggart informed me that he had seen it more plentifully, eastwards. The habit is diffuse; and Mr. Bennett refers it to var. aristatum Bab. Man. ed. 3, 153 (1851), English Botany, t. 2784 (non G. aristatum L.). Antennaria dioica Gaertn. Plants agreeing with the descrip- tion of var. piedicellata B. White, Scottish Naturalist, 1886, p. 323, were gathered on a hillock north-east of Garth Castle. Crepis mollis Aschers. (succiscefolia Tausch). Fine and plentiful, just north of Garth Castle; very local in Glen Lyon, about four miles above Fortingal. No definite station in the county was known to Dr. White ; but I understand that it has been gathered near KilHn by Messrs. H. and J. Groves. Taraxacum spectabile Dahlst. Meall Garbh, Creag Mohr, &c. ; apparently quite common on the mountains. Hteraciuvi anglicum Fr. The type and H. iricuvi Fr. grow by the Lyon at several spots, though neither is abundant. H. sinuans F. J. Hanb. Sparingly by the stream at the foot of Fin Glen, north-east of Ben Lawers, at 1600 ft. ; it ascends to nearly 3000 ft. on Ben Lawers itself, but is rare in this district. H. nigrescens Willd. Typical, but extremely scarce, by the Inverinain Burn, below Meall Garbh, between 1800 and 2000 ft. H. submurorum Lindeb. Specimens agreeing best with this were found at 1700 ft. by the Allt Odhar stream, north-west of Fortingal, among shaded rocks. Heads not quite so black as in 166 THE JOURNAL OF BOTANY typical examples ; but the rather low elevation may account for this, and I cannot refer them to anything else. H. ruhicundum F. J. Hanb. Stream-sides in Fin Glen, between 1600 and 2000 ft., very scarce; differing from type in the ligules being very ciliate and the styles very dark, but Eev. E. F. Linton agrees with me that it belongs to this species. Only recorded from Killin, in Perthshire. H. argcnteum Fr. Kocks by the Lyon, two miles below Fortingal, as well as higher up the river. H. Sommerfeltii Lindeb. Rocks at the east end of the Stuich- an-lochan, Ben Lawers, from 2700 to 3000 ft. ; the dark-styled form. H. i^seudonosmoides Dahlst. Locally frequent by the Lyon, and on rocks above it. H. callistojjhyllum F. J. Hanb. Typical, but very rare, l^y the stream in Fin Glen, at 1600 to 2000 ft. H. sylvaticum Gouan var. micracladium Dahlst. Allt Odhar, at 1700 ft. ; scarce. H. variicolor Dahlst. Not uncommon ; but both Mr. Linton and I fand it hard to draw a clear line between this and H. rotun- datum Kit. The leaves are, as a rule, more elongated ; when in shade, their blotches tend to disappear. Keltney Burn ; Allt Odhar ; streams below Meall Garbh (north side) ; rocks above the road from Bridge of Lyon to Fearnan. In this neighbourhood its range is from 2000 down to 1400 ft., with one exceptional station at about 400 ft. ■■'H. subnlatidens Dahlst. Mr. Linton concurs in referring to this some luxuriant, shade-grown specimens, with remarkably long foliage, found by the Allt Odhar, at 1700 ft. H. rivale F. J. Hanb. Stream-sides, Fin Glen and Allt Odhar, at 1700 ft. H. sagittatum F. J. Hanb. var. suhhirtum^ . R. Linton. Fin Glen ; Allt Odhar ; Inverinain Burn ; by the Lyon, near Invervar. Locally frequent, ascending to 2000 ft. H. euprepes F. J. Hanb., type. An alpine black-headed form occurs on cliffs at the east end of the Stuich-an-lochan, Ben Lawers, at nearly 3000 ft. — Var. glabra turn Linton. In many places by the Lyon, often very luxuriant, with lower peduncles up to nine inches long ; near Garth Castle ; Fin Glen ; eastern cliffs of the Stuich-an-lochan, Ben Lawers. From 300 to 2700 ft. -H. farrense F. J. Hanb. One fine plant, in a shady ravine 300 yards north of Garth Castle. U. custales Linton. Fin Glen and Inverinain Burn, between 1600 and 2000 ft. H. ccBsiovmrorum Lindeb. Allt Odhar, and by the Lyon, near Invervar, at about 700 ft. H. dissimile Lindeb. Rocks above the road from Bridge of Lyon to Fearnan (1400 ft.) ; Allt Odhar (1700 ft.) ; the usual Scottish form. H. Deivari Bosw. Near the Keltney Burn, &c. H. gothicum Fr. Abundant on a grassy, bushy hillock near NOTES ON SOME PLANTS OF MID-PERTH 167 Garth Castle, often as the forma latifolia W. E. Linton. An abnormal form of the type with sooty styles was collected on slopes above the Lyon, about six miles above Fortingal. ■'H. sparsifoliuvi Lindeb. This name is suggested by Mr. Linton for a yellow-styled hawkweed found on low rocks at the north-east end of Loch Tummel. It closely resembles one from near Inchrory, Banffshire, which his brother placed here, and under the type, in 1905 ; also (allowing for difference of situation) some much taller specimens from the Shee Water, two miles below Spital of Glen Shee, 1892. I think that all of them belong to this species. H. strictum Fr. Near the Keltney Burn, and in two or three places by the Lyon, but not plentiful. H. reticulatnm Lindeb. Frequent and often remarkably fine in Glen Lyon ; sometimes the leaves are untypically efloccose beneath. H. coryinbosum Fr. var. salicifolium (Lindeb.). Keltney Burn, and a little above Bridge of Lyon. Armeria maritima Willd. A very fine form of this, which I believe to be Syme's var. planifolia, grows in wet ground on the north side of Creag Mohr, from 2200 to 2500 ft. Mimulus moschatus Douglas. Well established in a muddy ditch about five miles above Fortingal, Glen Lyon. Euphrasia gracilis Fr. A form with white (instead of blue or reddish) flowers was noticed on a dry bank near the road, two miles west of Lawers Inn. Fine E. Bostkoviana Hayne {E. offici- nalis L., vera ?) is abundant in the low ground. BhinantJms Drummond-Hayi Druce. Very local on the northern slopes of Creag Mohr, at 2200 ft. Melampyrum pratense L. var. ■■'ericetorinn Oliver? Bushy hillock east of Garth Castle, in plenty. The name was given to it by Mr. Salmon, and it seems to agree quite well with the description in Phytologist, 1852, p. 678 ; but I have not seen Irish specimens. In the stronger plants the bracts are decidedly toothed, not entire, as in var. montanum. I have found precisely the same thing at Wybunbury Bog, Cheshire. Mentha piperita L. Established in bushy ground near the Lyon, three-quarters of a mile below Fortingal. Mr. Salmon determined it as a. officinalis (Hull), our normal form. Polygonum aquale Lindman. Common in Glen Lyon, below Fortingal ; named by Dr. Moss. It varied a good deal ; one of the forms agrees with the plant figured by Syme as P. aviculare var. arenastrum. Betida alba L. x pubescens Ehrh. var. microphylla E. S. Mar- shall. Two trees of this hybrid combination, which seems to be new, grew with the parents by the Allt Coire Pheiginn, west of Garth Castle, at 700 ft., both being good intermediates {B. pube- scens var. microphylla is frequent in the district). Dr. Moss and I agreed in this determination. Ahius glutinosa Gaertn. The only form observed is var. microcarpa Eouy {teste Moss). Salix caprea L., var. sphacelata Wahlenb, Allt Coire Pheiginn 168 THE JOURNAL OP BOTANY (700 ft.) ; stream descending from the north side of Meall Garbh (1600 ft.) ; both certified by Dr. Moss. Probably frequent in glens of the Highlands ; I am inclined to think it a good average variety. S. nigricans x 2:)hylicifolia. An excellent intermediate was found by Dr. Moss and myself on Ben Lawers, very little belov^r 3000 ft. Taxus baccata L. Not native ; but two trees of considerable age grow on the cliffs above Fortingal, at 700 and 1100 ft., doubtless bird-sown from the very ancient yew in the churchyard. Habenaria conopsca Benth. A variation with deep claret- coloured flowers was noticed in the Lyon valley, below Fortingal. Polygonatum verticillatum All. Apparently new for the Breadalbane district ; shown to us by Mr. Haggart in a rocky ravine, associated with Gonvallaria majalis L. Juncus biglumis L. Meall Gruaidh (or Greigh) ; head of Fin Glen ; north side of Creag Mohr. EriopJwrum latifolium Hoppe. Hill-bog, west of the Lawers Burn, at 1200 ft. ; extending only over a very limited area. Kobresia biimrtita Dalla Torre (ccm'cena Willd.). Occasionally reaches 3000 ft. on the Ben Lawers range ; it occurs on the north side of Creag Mohr. Carex aquatilis Wahlenb. A tall, slender form was met with in profusion near the west end of Loch Tummel. — Var. virescens And. By the Lyon, about six miles above Fortingal ; confirmed by Mr. Bennett. C. atrofusca Schkuhr. We obtained this in two corries (about a mile apart) of a mountain, south of Glen Lyon, for which it was not known before ; one fruiting stem measures sixteen inches, and several were fully a foot high. The leaves have a decidedly glaucous hue, which I have not seen mentioned ; Mr. Druce is evidently right in placing it between C. atrata L. and C. rarijiora Sm. We found the altitude to vary from 2300 to about 3000 ft. C. fulva X CEderi (type). Northern shore of Loch Tummel, in small quantity ; the hybrid of C. fiUva with C. (Ederi, subsp. cedocarpa And. is quite common, as usual. "^'C ^7^-/Zato Huds. (rosira to Stokes) x vesicaria. Marshes at the head of Loch Tummel, showing considerable vai'iation. Agrostis nigra With. Native in grassy ground by the river at Fortingal ; we also gathered it in Mr. Druce's station near Lawers Inn, but less dark-flowered. Poa nemoralis L. An intensely glaucous plant, apparently var. ccssia Gaud., is abundant on the upper cliffs above Fortingal, from 1000 to 1200 ft. '-'Glyceria declinata Br6b. Muddy ground near Fortingal Hotel ; also close to Lawers Inn. Equisetum arvense L. What I know as ' var. nemorosum Braun ' grows in damp shade at Fortingal, and by the road from Bridge of Lyon to Fearnan. ■'•E. arvense x Uviosum {E. litorale Kiihlewein). South shore of Loch Tummel, near the shooting-lodge ; also at the north-west end. In bo^h cases the parents were associated with it ; the A NEW CHIMAPHILA PROM SAN DOMINGO 169 terminal cones, by no means freely produced, are very small and abortive. New for Scotland ; but probably overlooked. E. variegatum Schleieh., a. arenarium Newm. A few plants on Meall Garbh, at 2600 ft., and by Loch Tummel, where it is associated with Myrica gale and Schcenus ferrugineus. A NEW CHIMAPHILA FEOM SAN DOMINGO. By Sidney F. Blake, A.M. While working recently on the variations of the widely dis- tributed Winter green known as Chwia2:)hila umbellata (h.) Burt., ■■'• I was struck by the marked difference from all other specimens in the British Museum shown by those collected in San Domingo by Baron Tiirckheim, and referred on the label to this species. Although the plant has long been known to range over a large part of temperate Europe and Asia, nearly all of North America from Canada to the mountains of Mexico and Guatemala, and Japan, its presence in the West Indies has only recently been indicated by Urban (Symb. Antill. v. 453 (1908)), on the basis of specimens collected long ago by Eggers ; but a careful examination of Tiirckheim 's specimens, which are in fresh flower, and of duplicates at Kew of Eggers' plant, which is in old fruit, shows that they represent a decidedly distinct species. Not only are the leaves much smaller and fewer-toothed, but the peduncle and pedicels, densely dotted with glandular atoms in all the forms of G. livihellata, are here perfectly glabrous, while the filaments, widened and short ciliate near the middle in G. umbellata, in the San Domingo plant are enlarged nearer the base and are quite glabrous. In view of these marked characters the San Domingo plant seems best considered an endemic species which may bear the name / -: Chimaphila domingensis. Planta humilis 10-15 cm. alta ; foliis cuneato-obovatis obtusiusculis subsessilibus 1-8-2-5 cm. longis 6'5-8 mm. latis dentibus 2-5-jugis hebetibus ; pedunculis (et pediculis) glaberrimis 4-6 cm. longis ; floribus 3-5 subumbel- latis ca. 11 mm. latis ; sepalis orbicularibus glaberrimis margine minute eroso non ciliato ; petalis suborbicularibus minute irregu- lariter erosis (purpureis fide Eggers (apud Urban), sed in specim. Tiirckheimianis certissime albidis) ; antheris violaceis ; filamentis basi ovate ampliatis glaberrimis. Santo Domingo : pine woods, alt. 2200 m., Valle Nuevo, near Constanzu, August, 1910, Tiirckheim, 3434 (type in Brit. Mus.) ; among stones, alt. 2270 m., Valle Nuevo, May 30th, 1887, Eggers, 2269 (hb. Kew.). I am indebted to Dr. Eendle for permission to dissect a flower from the Tiirckheim specimens in the British Museum. ^^ * This combination has been universally quoted from Nutt. Gen. i. 247 (1818), but can hardly be considered made there, as no synonym is cited; however, it had been properly made the previous year by Barton (Veg. Mat. Med. U.S. i. t. 17, t. 1 (1817) ), as indicated by Gray, Syn. Fl. U.S. (1818). 170 THE MANX SAND-DUNE FLOEA. By J. W. Hartley and J. A. Wheldon, F.L.S. Early in June of the present year we spent half a day on the sand-dunes of the north coast of the Isle of Man. Our chief object was the collection of lichens, but we were also desirous of comparing the vegetation of these dunes with the flora of the similar but more extensive eolian sands of the coasts of Lancashire and Cheshire. The season was too early for a thorough investi- gation of the higher plants, and we are sensible that the lists given are far from complete. Sufficient, however, was noted to show that these dunes are of a totally different type from those of Southport, Wallasey, or St. Anne's. The area of blown sand is not extensive, commencing near Jurby and extending in a narrow belt towards Point of Ayre. Inland the dunes overlie and merge into the boulder clay, which eventually rises into a range of low hills following the coast-line. At Point of Ayre there is a wide expanse of barren heath, where the blown sand only partially covers extensive beds of shingle, and does not form dunes properly so called. In many places the sand has been blown away except where- held by the roots of ling and heather, exposing the shingle or boulder clay beneath. The moorland then bears a not distant resemblance to the peat-hags of our moorland summits, only the peaty bases of the stools are composed of sand instead of peat. This ground looked interesting, and appeared to be especially rich in CladonicB, which were fine and well-developed. Unfortunately we arrived too late to give it an adequate examination, and dark- ness coming on we were reluctantly compelled to leave the exploration of this Point of Ayre heathland for a future excursion. As compared with that of Southport, the flora of these dunes is poor and uninteresting. The species are fewer and of a less specialized dune-type. The differences which immediately strike the observer may be briefly summarised : — (1) The dunes are comparatively flat, and provide less variety of aspect than those of the mainland. The absence of high dunes and intersecting deep hollows, besides affecting the flora, diminishes the picturesqueness of their appearance. (2) The whole of the coast we examined, both dunes and heathland, was very dry. There are no wet " slacks." The flora is therefore more uniformly mesophytic and xerophytic than at Southport, the rich bryological and hydrophytic associations of the latter locality not being represented. (3) The absence of Salix repens, a dominant species of the Lancashire dunes, enhances the general flatness ; the secondary dunes built up by this species adding greatly to the diversity of the surface when it is present. (4) The calciphilous species are less prominent, the restricted calciphile species being absent. (5) The larger fungi are very scarce, much more so than on the damper dunes of the mainland, where, from May onwards, THE MANX SAND-DUNE FLORA 171 they may always be seen in profusion. Not a single species was noticed on this occasion, and only one or two on a previous visit, some years ago, in September. (6) The rapid transition from the sand-dune vegetation to that of heathland, with the presence of Pteris, Callitna, Erica, and Ulex, all of which are either excessively rare or absent from the Lancashire dunes, is especially noteworthy. The causes which bring about these differences are apparently twofold. The dominant factor is probably the great depth of the subsoil water, due perhaps to the sand overlying pebble beds derived from the boulder clay. We were informed that at Point of Ayre it was possible to dig a depth of some seventeen feet without reaching water. At Southport, on the contrary, it lies only a few inches beneath the surface of the deeper hollows, and often stands in pools. Another factor, which probably strongly affects the vegetation, is the apparently smaller lime content of the soil. Shells are less plentiful in the sand, most of which is probably derived from the boulder clay of the north of the island. The bulk of sand is also smaller than that delivered to the estuarine dunes of the mainland. Owing to this deficiency of material, or to the steeper inclination of the land, the dunes are less extensive and of slower growth, and appear to go through a constant alternation of growth and demolition. The high boulder clay banks in the vicinity prevent the spread of blown sand to any considerable distance inland, and the material is not supplied in sufficient quantity to enable the dunes to make any marked advance on the seaward margin. Here in several places there is abundant evidence of recent erosion. The poverty of species will be evident on perusal of the following list, which, whilst doubtless very incomplete, probably includes the majority of the plants found at the several points visited. The List of Plants of the Isle of Man, published by the Kev. S. A. Kermode, does not help much, as localities are rarely quoted, but some of those he enumerated may possibly have been from this area. If so, they are certainly not obtrusively evident. Amongst those he names which are likely to occur are Glaucium luteum, Crambe maritivia, Trifolium striatum, T. arvcnse, Anthyllis vuliieraria, Carlina vulgaris, Leontodon hirtus, Erythrcza Cen- taurium, Etqjhorbia Paralias, E. portlandica, and Sedum acre. Most of these are abundant on the drier Lancashire dunes, and there seems to be no adequate reason for their absence here, if they really are absent. About Jurby and Lhane several well-marked zones of vege- tation could be discerned. I. The Strand Association. — This in some places is developed on shingle, in others on sand. It forms an almost continuous band dominated by Atriplex, amongst which we were able to distinguish (although not yet flowering) A. hastata Ij., A. Bahing- tonii Woods, and A. laciniata L. Other plants observed were Cakile maritima Scop., Glaux maritivia L., Arenaria i^eploides L., and Salsola Kali L. On pebbles where the ground was shingly 172 THE JOURNAL OF BOTANY a few minute lichens develope. On a few pebbles brought home for examination we found Acarospora pruinosa Jatta f. nuda Nyl. ex Lamy., A. sniaragdula Ach., and Bhizocarpon confervoides DC. f . fuscescens Leight and f . dispersa Leight. We did not observe at any of the points visited a fringing zone of Agropyron junceuni, such as may be seen between Ainsdale and Birkdale, but the species occurs sparingly on the coast. II. The Marram Association. — The dunes rise rather abruptly from the strand, in some parts in a continuous slope, in others with a low steep scarp. In the latter case the freely exposed roots of Marram grass show that erosion by tide or wind has taken place. This grass is the dominant plant, and the slightly undulating dunes it forms are neither as lofty nor as much cut into hills and valleys as those of the mainland. After reaching their maximum height, which is low as compared with the dunes between Formby and Southport, they fall in a gentle slope towards the land, forming a depression or shallow valley, which roughly and with some interruptions follows the contour of the coast-line. There are a few deeper hollows in which water probably stands for a short time in wet seasons, but there are no permanently wet "slacks." In these hollows a number of mosses and lichens carpet the ground, but none of them are of a distinctly hydro- philous type. The following species of this association were recognizable at this early season of the year : — Ammophila arenaria Link, dominant. Agropyron junceum Beauv. Taraxacum officinale Weber. Festuca rubra (aggr.). T. ohliquum Dahlst. Senecio vulgaris L. Carex arenaria. The above plants are the first to appear ; further from the littoral zone they all still persist, with the addition of the following : — Arenaria serpyllifoUa L. var. Valerianella olitoria Poll. macrocarpa Lloyd. Galiiim verum L. var. viariti- Cerastium tetandrum Curt. mum DC. C. semidecandrum L. Polygomim Itaii Bab. Erodium maritimum L'Herit. Phleum arenariiim L. E. cicutarium var. glutinosum Aira caryoph^jllea L. Clav. Bromus hordeaceiis var. lepto- Eryngium maritimum L. stachys Beck. A single lichen was noted on dead stools of Marram grass, viz. Cladonia fimbriata Fr. f. exilis (Ach.). III. The Marram loith Bracken Association. — In the lower part of the main depression, Pteris aquilina begins to appear amongst the Marram grass, and soon forms a well-marked zone, sharing the ground with the grass, and in places becoming the dominant species. Here many of the plants mentioned in the preceding group continue to flourish. Lichens and mosses carpet the ground in profusion, thriving where the sand is enriched in humus by the decay of the more luxuriant vegetation. Cladonia alcicornis forms light-coloured patches, and together with other THE MANX SAND-DUNE FLORA 173 CladonicB is much more abundant than on the Lancashire dunes. The species noted were as follows : — Pteris aquilina L., dominant. Ammopliila arenaria Link ,, Brassica monensis Huds. (rare). Viola ericetorum Schi-ad. Arenaria Lloydii Jord. Cerastium tetrandum Curt. C. semidecandrum L. C. viscosum L. Polygala oxyptera Reichb. Erodium cicutarium var. gluti- nosum Clav. Ononis maritima Dum. Trifolium duhium Sibth. Lohis corniculatus L. var. crassi- folius Pers. Sedum anglicum Huds. Galium verum L. Valerianella olitoria Poll. Bellis i:)eren7iis L. Taraxacum officinale Weber. Matricaria inodora L. Jasione montana L. Calystegia Soldanella Br. Myosotis collina Hoifm. Thymus Serpyllum L. Plantago Coronop^is L. P. lanceolata L. Carex arenaria L. Aira caryophyllea L. A . pracox L. Anthoxanthum odoratum L. Agrostis alba L, Festuca ovina L. F. rubra (aggr.). Bro7nus hordeaceus L. Botrychiicm Lunaria Sw. Bryophyta. Ceratodon purp^ireus Brid. Dicranum scoparium Hedw. D. scoparium var. orthophyllum Brid. Barbula convoluta Hedw. B. Hornschuckiana Schultz. Tortula ruralis Ehrh. T. ruraliformis Dixon. Bryum pendulum Schimp. B. capillar e L. B. roseum Schreb. Brachythecium albicans B. & S. Brachythecium velutinum B. & S. B. purum Dixon. Camptothecium lutescens B. & S. Hypnum cupressiforme L. H. cupressiforme L. var. tecto- rum Brid. Hylocomium splendens B. & S. H. splendens var. gracilius Boul. H. squarrosum B. & S. H. triquetrum B. & S. LiCHENES. Peltigera canina Hoffm. Cladonia alcicornis Floerke. C. pyxidata Fr. var. pocillum Fr. C. chlorophcea Flk. C. fimbriata Fr. f. exilis (Ach.). Cladonia fimbriata var. conista Nyl. C pityrea Flk. G. furcata Hoff m. C. pungens Flk. Bacidia muscorum Mudd. In addition to the above, the following three lichens were found growing on pieces of weathered old leather lying on the dunes : — Lecanora Hageni (Ach.). Buellia phacodes Koerb. Binodina exigua Gray f. demissa Stiz. IV. The Marram luith Heather Association. — As the landward border of the preceding association is reached, the bracken begins to fail, and intermixed with it are a few plants of Galluna vulgaris. Further on this becomes more abundant, and is accompanied by a small quantity of Erica cinerea and Bosa spinosissima. In 174 THE JOURNAL OF BOTANY conjunction with a diminished quantity of Marram grass these plants dominate the rest of the ground. This slopes rapidly upwards, and soon becomes ordinary heathland, characterized here by the occurrence of Ulex and Calluna. The Ulex is the last plant to enter the dune formation, and only appears at about the point wliere the Man-am disappears. In the list of plants of this association given below, those only are tabulated which occur on what may be termed the dune-heath, viz. that portion in which the ericetal species are accompanied by Ammophila arenaria. Where the Marram disappears and furze becomes a common associate of the heather, which occurs as the sand thins out over the boulder clay, the flora belongs to a different formation. Time did not allow of a careful examination of this heathy tract, which has no equivalent on the Lancashire coast, where the sand-hills gradually merge into dune-pasture and cultivated ground. The following species, all noted within the range of the Marram grass, belong to this association : — Calluna vulgaris, dominant. Senecio Jacohcea L. Pteris aquilma, ,, S. sylvaticus L. Ammophila arenaria, „ Arctium {mimis Bernh.?). Bosa spinosissima, siihdoimna,nt. Centaurea nigra L. Erica cinerea, Rammculus hulbosus L. Erophila verna E. Meyer. Viola ericetorum Schrad. Polygala oxyptera Keichb. P. serpyllacea Weihe. Cerastiiim viscosum L. C. vulgatum L. Geranium molle L. Ononis repens L. Meclicago lupulina L. TrifoUum repens L. Lotus corniculatus L. Vicia angustifolia L. Conopodium denudatum Koch. Anthriscus vulgaris Bernh. Bellis perennis L. Leontodon autumnalis L. Campanula rotundifolia L. Lycopsis arvensis L. Veronica officinalis L. Euphrasia curta Wettst. Thymus Serpyllum L. Plantago lanceolata L. Bumex crispus L. B. Acetosella L. Luzula mnltiflora DC. Anthoxantlmm odoratam L. Aira caryophyllea L. A. prcecox L. Cynosurus cristatus L, Dactylis glomerata L, Lolium perenne L. Festuca ovina L. Bryophyta. Polytrichum pmiperimim Willd. Campylopus fragilis B. & S. Dicranum scoparium Hedw. D. scoparium var. orthophyllum Brid. Ceratodon purpureus Brid. Tortula mralis Ehrh. Ftimaria hygrometrica Sibth. Webera niitans Hedw. Bryum inclinatiim Bland. de- Climacium dendroides var. pauperatum Boul. Brachythecium purum Dixon. Hypnum Schreberi Willd. H. cupressiforme L. H. ciqjressiforme var. ericetoriim B. & S. Hylocomium squarrosum B. & S. H. triquetrum B. & S. Ptilidium ciliare Hampe. Frullania Tamarisci Dum. THE ORIGIN OP SPECIES 175 LiCHENES. Peltigera canina Hoffm. Gladonia fivibriata var. conista P. rufescens Hoffm. Nyl, P. phy socles Ach. C. gracilis Hoffm. Cetraria aculeata Fr. C.furcata Hoffm. Gladonia pyxidata Fr. C. furcata var. corymhosa Nyl. C. pyxidata var. pocillum Fr. C. pungens Nyl. C. chlorophcBa Flk. Cladina sylvatica Nyl. C.fimhriata Fr. f. exilis (Ach.). C. uncialis Nyl. C fimbriata var. tuhcBformis Bilivibia ligniaria Massal. Fr. Lecidea idigiiiosa Ach. THE OEIGIN OF SPECIES. [The following is an abstract of a paper on " The Origin of Species by Crossing " read at a meeting of the Linnean Society on the 19th of February, by Dr. J. P. Lotsy, of Haarlem. The paper was illustrated by diagrams, lantern-slides, and dried specimens.] We have in all questions of evolution to gather our facts from individuals, because species as well as varieties are abstractions, not realities. Nobody is able to show you a species or a variety ; all he can do is to show you one or more individuals which he believes to belong to the species or variety under discussion. Of individuals we know two kinds : homozygotes and hetero- zygotes. The first are stable; the latter segregate, earlier or later, into new homozygotes. The offspring of a homozygote is identi- cal with its parent, with the exception of mere temporary non- transmittable modifications. If this be true, selection in the pro- geny of a definite homozygote can have no effect. That it has no effect has been proved by Johannsen. A homozygote con- sequently is absolutely stable and produces offspring which is genetically identical with it. Yet not all homozygotes are the same, there are many different kinds of homozygotes : homo- zygote beans, homozygote Antirrhinums, &c. All these different kinds of homozygotes we may call with Johannsen genotypes, because they differ in genetical constitu- tion, and we can then say that the world is populated — with the exception of heterozygotes — by a large number of sharply-defined absolutely stable genotypes. Under such conditions evolution may well seem impossible ; fortunately, the behaviour of the heterozygotes shows us that it is quite possible. A careful study of the descendants of a heterozygote shows us that it segregates in the next or later generations in a number of individuals, part of which are heterozygous, but part of which are homozygous, and that these homozygotes belong to different genotypes. A heterozygote consequently gives birth to a smaller or larger number of different genotypes. 176 THE JOURNAL OP BOTANY By carefully watching therefore a heterozygous individual we see the origin of genotypes. The next step is thus to produce at will these genotypes — originating heterozygotes. This we can do by crossing two individuals belonging to different genotypes. The next question is : Do all heterozygotes obtained by crossing segregate and thus give rise to different genotypes ? Until very recently it was believed that only heterozygotes obtained by crossing so-called varieties did segregate, while crosses between so-called species were said to give a stable offspring. Evidence is rapidly accumulating that this is not true as far as the latter is concerned, that species-hybrids also segregate. I therefore claim that the origin of genotypes by crossing is of much wider application than was formerly supposed, that perhaps it is of universal application. I further claim that the genotypes are the real, long sought-for units of the natural system, and I propose that in future the term species be limited to them, in accordance with the view held by systematists for ages that " species " is the proper term for the units of the natural system. [The question of Progressive Mutations, based upon Prof. H. de Vries's work upon (Enothera Laviarckiana, was then discussed.] The chief question is whether (2/. Lamarckiana is a pure geno- type, because only if the purity of type is beyond any possibility of doubt is there good reason to explain the throwing of deviating types as due to new formation of factors ; in any other case these can be explained as the result of new combinations of factors already present in the plant throwing them. Heribert Nilsson has, in the opinion of the speaker, undoubtedly shown that CE. Lamarckiana is no pure genotype, and consequently cannot serve as a reliable basis for the study of the origin of mutants. Nor did the speaker know of any other case in which progressive mutation from a pure genotype has been proved. [Mutation by loss of Factors was next considered. It was submitted that but one thing is proved, viz. ; that the real units of the living kingdom are genotypes ; that such genotypes can, under proper precautions, be kept pure for an indefinite time ; and that there is no certain evidence that they can be changed in any other way than by crossing.] What then is the reason of the apiparent variability of a species in the Linnean sense ? In the first place, the fact is that a Linnean species is a collection of independent stable Jordanian species. Indeed, as Bateson says ; " Between Jordan with his 200 odd species for Erophila and Grenier and Godron with one, there is no hesitation possible : Jordan's view .... is at least a view of natural facts, whereas the collective species is a mere abstrac- tion." The Linnean species, indeed, has been a snare, and if we, as Darwin did, consider it as a unit, the small species contained in it must naturally appear to be deviations of the type — in other words, varieties. If, then, one further sees that between Linnean species which one considers to be units, transitional forms exist, it is perfectly logical to build up a theory that one species can THE ORIGIN OF SPECIES 177 change to another hy means of its variabiHty. Besides, the illusion of variability is not created by the presence of these small species alone. As Bateson says : " When this variability is sorted out, and is seen to be in part a result of hybridization, in part a consequence of the persistence of hybrids by partheno- genetic reproduction, a polymorphism due to the continued presence of individuals representing various combinations of Mendelian allelomorphs, partly also the transient eifect of altera- tion in external circumstances, we see how cautious we must be in drawing inferences as to the indefiniteness of specific limits from a bare knowledge that intermediates exist." The author expressed his firm conviction, as explained before, that no trans- mittable variation exists, and that all apparent variability is due to an original cross. Finally, the author proceeded to the origin of species before sexual reproduction took place. He laid stress on the fact that this of course is mere speculation. As such he offered the following points : — (1) If a species is a perfectly stable genotype, reproducing faithfully its kind for ever, unless crossing interferes, all differences between the individuals belonging to a geno- type must be non- inheritable modifications. (2) Inheritable variability does not exist, with the always possible exception of mutation through loss of factors. All that has been described as variability is the result of vegetative or generative segregation of heterozygotes. (3) No inheritance of acquired characters occurs. The total of the inheritable factors now found among higher organisms must have been present in the total of the " urorganisms," each of which, however, possessed but a small number. [At the meeting of the same Society on June 4, the Eev. George Henslow gave an address on "Darwin's Alternative Explanation of the Origin of Species, without the Means of Natural Selection," of which the following abstract was supplied.] The cause of variation, always " changed conditions of Life," with " Definite or Indefinite Results " {Variation, £c. vol. ii. p. 272, 1868). Natural Selection is not a cause : e. g. Mivart {Origin, dx. 6th ed. ch, vii.). Also, E. W. Hutton, who says : — " Having Natural Selection to be a true cause, and one that largely explained the Origin of Species from Varieties, by causing a gradual diver- gence of character, &c." {Darioinism and Lamarckism, p. 38, 1899). No necessity for the words " several generations " for giving rise to a variation, as plants vary at once ; but they are required for fixing them so as to be heredity under any conditions. "Species" and "Variety" are terms representing the varying amounts of change necessary for adaptation. Hence there need be no intermediate forms. First reference to " definite action," &c., as a cause {Origin, &c. Journal op Botany. — Vol. 52. [July, 1914.] r 178 THE JOURNAL OF BOTANY 1st ed. End of Introduction, 1859). His original view {Origin, dc. pp. 11, 12) not strongly emphasized until 1868 (Variation, etc., and Origin, d'C. 6th ed. 1878). Reason for delay given to Wagner {Ltfe, it-c. vol. iii. p. 158; Oct. 1876), viz :— " I could find little evidence of the direct action of the environment." " Definite variation " leads to " permanent modification of structure," i. e. inheritance of acquired characters {Origin, itc. 6th ed. p. 421, 1878). These may become only relatively, or even absolutely, stable under all conditions of life. Indefinite variations, also caused by " direct action of changed conditions of life." They consist of a supposed mixture of indi- viduals, the minority possessing " favourable " {i. e. adaptive), the ynajority, "injurious" {i.e. inadaptive) variations. Such is the Theory of Natural Selection as described {Origin, dc. 6th ed. p. 63, 1878). Natural Selection is not required, for the majority die by "fortuitous destruction" {Origin, £c. 6th ed. pp. 53, 54, 59-89), e. g. Sir E. Ray Lankester says, one out of a million eggs of an oyster may survive, i. e. per chance (Darwin's word). Contra, F. Buckland says of young oysters, transferred to new localities, that within two months they begin to assume the " native characters " {Variation, dc. vol. ii. pp. 280, 281). In illustration, Mr. Henslow adduced the following examples : — Mesophytic plants becoming aquatic and vice versa, e. g. Water Crowfoot. E. g. Monocotyledons originated from aquatic Dicoty- ledons. Mesophytic plants becoming Xerophytic and vice versa by means of water, e. g. Restharrow, &c. Cultivated plants originating, by prepared soil, from wild plants, and their rever- sion, e. g. root crops. Depauperised and dwarfed plants, recog- nized as specific ; through drought, submergence, parasitism, and saprophytism. Darwin's final charge against scientists for their misrepresen- tations : — " Great is the power of steady misrepresentation; but the history of science shows that, fortunately, this power does not long endure " {Origin, &c. 6th ed. p. 421). It has lasted for fifty-five years (1859-1914). NOTES ON DR. FOCKE'S BUBI EUROPE I (1914).- By the Rev. W. Moyle Rogers, F.L.S. Biobi Europcei is the latest product of Dr. Focke's unrivalled work on the brambles of the world, and of Europe in particular. Here we find in great detail his maturest views on our British Rubi in correlation with those known on the continent of Europe. We can hardly be too grateful to him for the untiring labour * Bibliotheca Botanica. Herausgegeben von Prof. Dr. Chr. Luerssen Danzig-Zoppot. Heft 83. Wilhelm Olbers Focke : Species Kuborum. Mono- graphise generis Rubi Prodromqs. Pars iii. Stuttgart. 191-4. NOTES ON DR. POCKE'S RUBI EUROPiEI 179 evident throughout. Some criticism of the conclusions reached is, however, inevitahle. And naturally those whose opportunities of study have been more concentrated through a long series of years on the forms occurring in any one country, or group of neighbouring countries, may still doubt whether the last word has been said in every case. As will be seen by anyone turning over the pages (in the earlier groups especially), Dr. Focke finally assents to most of our names, though he often groups them differently, as was to be expected in so elaborate a classification as that now formulated by him. To this difference of grouping only occasional reference will be found in the following notes. The few illustrations from photographs which occur here and there — however unsatisfying in some respects, as they must always be — are yet of real value as reproducing exactly both leaves and detailed outline of flowering branches. The first three groups of the Fruticosi section of Bubus in our London Catalogue, ed. x. {Siiberecti, Suhrhamnifolii, and BJiamni- folii) contain twenty-four numbered " species." Of these, four (castrensis, incnrvatus, diirescens, and lacustris) are, so far as is at present known, endemic in Britain, and so necessarily retain our names in Buhi Europai ; as do eighteen out of the remaining twenty, leaving only two of the twenty-four — B. integribasis and B. LindUianus — in doubt. Of these, B. integribasis must now apparently be either struck out of our list, or provisionally, as a matter of convenience for reference, changed from " B. integri- basis P. J. Mlill. ? " (as it now stands) to B. integribasis Rogers {non P. J. M.) ; as Dr. Focke, who first suggested the name for our plant, has, after considerable previous hesitation, distinctly- disallowed it. He says of it : " In planta Britannica [B. integri- basis Eogers) foliola potius obovata, aculei pauUo longiores et robustiores sunt. Stamina stylos superant. Petala roseola. B. cceresiensi [Sudre et Gravet] sine dubio magis aflfinis quam B. integribasi." And he adds, " Im siidhchen England." Our plant, as thus distinguished, is locally abundant throughout South-east Dorset and along the New Forest border in South Hants, extending in that direction to Woolmer Forest, North Hants. I have also seen it on Tooting Common, Surrey. It must therefore have a permanent place in our list, and ultimately a new name. In the case of B. Lindleianus, Dr. Focke now writes : " B. vulgarem et B. Lindleyanum olim (Syn. Rub. Germ.) dis- tinguere conatus sum, sed revera omnino confluunt et limites naturales non existunt." Whether in consequence of this we should change our name is no doubt a question for decision ; as to which, I must own, I feel no little difficulty. Lees's Lindlei- anus is one of our most widely distributed and strongly marked British brambles, with which I have long been very famihar. Of the German B. Lindleyanus (Syn. R. G. and Aschers et Graebn. Syn. Mitt. Fl.) I have seen no specimens ; but I have six sheets of " B. vulgaris Wh. & N.", collected by Dr. Focke at Minden (1871 and 1873), Rinteln (German, boreal.) 1872, and Bremen p 2 180 THE JOURNAL OP BOTANY 1889, and I could not put these together as one species ; nor does any single sheet seem to me identical with our plant, though the two Minden sheets are near it. The figure of R. vulgaris (tab. xiv. p. 38) in Eub. Germ, is also considerably unlike our R. Lindlei- anus. So I am still unwilling to surrender our name for Weihe & Nees' more aggregate one. As to R. Rogersii Linton, there is happily no longer room for uncertainty. It is now accepted by Dr. Focke as a good " species," and as (so far as is yet known) endemic in the British Isles, while his R. mnmohius takes subordinate rank after it, as " R. 'pUccito et a. Rogersii similis." In further differentiation of the German plant he adds, " foliola multo majora quam in R. Rogersii, vix plicata ; terminale in foliis quinatis cordato-suborbiculare vel cordato-ovatum ; infima breviter petiolulata. Rami fertiles, petioli pedunculique multo parcius aculeati quam in R. Rogersii. Flores majores quam in R. jiUcato et R. Rogersii. Petala incurva, i.e. e fundo patente ascendentia." Though thus distinct enough from our plant, it is obviously allied to it. The above notes refer only to numbered " species " in our first three groups. In addition to such species, we have eight well- marked forms in these groups. In the London Catalogue these are all entered as varieties, the plan of the Catalogue being to class all the plants admitted into it as either species or varieties. But it must not therefore be taken for granted that all these "varieties" (either here or elsewhere) are necessarily of equal value, i.e. as conspicuously distinct all from the species with which they are placed. I have never thought them so certainly, as I tried to show by dividing them into " sub-species " and " varieties " in my Handbook of British Ruhi. These more or less subordinate forms are not always easily traced in Ruhi Europcei. This is partly due to the author's comparatively slight acquaintance with some of them, only a few having been seen by him when he visited England in 1889 and 1894 ; while he has not always been kept duly supplied with dried specimens. Of those in our first three groups, however, Briggsianus is the only one which he seems not to have noticed. As to the rest, he suggests no important change or rearrangement, except in the case of Bertramii, which, apparently, he no longer accepts as British, though I can find no discrepancies between his fuller description and my short one in my Handbook of British Ruhi, p. 22, nor between our plant and " Slesvig " specimens received from Mr. Friderichsen. His R. 02)acus, however (a variety of R. nitidus in the London Catalogue, but a sub-species in my Handbook), he places as an independent but unnumbered form or species after his R. ammobius, with the note added : " R. nitidus subsp. opacus Rogers Handbook of British Ruhi, p. 23, ex mea sententia non differt, quamvis R. opacus R. plicato magis affinis videtur quam B. nitido." Thus far I have written as if our grouping fairly corresponded with that adopted in Ruhi Europai. But, as I have stated above, it does so only partially, even while we are considering the earlier NOTES ON DR. FOCKe's RUBI EUROP^ffil 181 species, as in the above notes. Dr. Focke's arrangement, as now elaborated by him for the Ruhi of the World, is really much more complicated than ours, and, as we advance beyond our first three groups, a detailed comparison between the two systems, and the attempt to keep a corresponding order in dealing with the plants, becomes increasingly difficult. From this point, therefore, I may content myself with briefer notes bearing almost exclusively on details connected with the specific and varietal names. Thus the Villicaules — a small intermediate group with us — is now divided by Dr. Focke between his more aggregate Bhanmifolii and Silvatici, and nothing need be here added to what may be found on its species and varieties in the 1905 volume of this Journal, p. 201. As regards our Discolores also, his treatment now is not materially changed from that found in his Aschers. et Graebn. Syn. Mitteleur. Fl., pp. 499-512, and is hardly such as to call for special comment here. When we come to Silvatici and Vestiti it is otherwise ; and changes are suggested of special interest to us. Our B. hesperms is no longer associated by Dr. Focke with his i?. myricce, but is placed by him in an independent position, as a numbered species, between B. Sprengelii and R. Arrhenii ; while B. myriccB is removed to a place among the Egregii, with the following explanation : " Prima, quae vidi, specimina comparavi cum B. myricce. meo {vide Rogers, Handbook, 1. c), sed, plantis iterum atque iterum examinatis, differentiae nunc graviores mihi apparent. B. myricce ex affinitate B. Silvatici removi." So B. myricce disappears from our list. I may add that though B. hesperius, exactly as described in the 1896 volume of this Journal, has so far been found only in Ireland, a very closely allied form, which need not perhaps claim a separate name, occurs near Bangor (Carnarvon) and in three or four English counties. It is unknown on the Continent, and Dr. Focke expressly separates from it the " B. hesperius, Piper, Erythrea v. p. 103," published two years later than the Irish plant. B. Colemanni Blox. Dr. Focke now places this between B. villicaulis and B. Sehneri ; but our position — last among Silvatici and close to Vestiti — seems preferable. He also says of it: " Verosimile mihi videtur B. Colemanni auctorum recentiorum ' speciem ' aggregatam esse. Plantae Bloxamii typicae originem hybridam {villicaulis x rculula?) suspicor." Such an opinion should not be lightly brushed aside. But opportunities of studying the living plant have greatly multiplied in recent years, and it proves to be very widely spread in England, east and west, from Yorkshire to Hants. It is also locally abundant, and, as a rule, quite remarkably constant. Bloxam's dried specimens of his Leicestershire plant have shorter and usually more roundish leaflets than I have seen from any other county ; but I cannot see that they are otherwise different. B. orthoclados Ley. In Aschers. et Gr. Syn. vi. 470 (1902) a new name {B. euchloos Focke) was substituted for Ley's name, a change not welcomed by Mr. Linton and me in our paper in the 182 THE JOURNAL OF BOTANY 1905 vol. of this Journal, p. 201. Ley's name is now restored for the British plant, and B. cuchloos separated from it and transferred to M. Gravet's Belgian plant, referred to on p. 47 of my Handbook. But a further suggestion is now made that both are probably of hybrid origin: the British as " plicatus x Sprengelii,'' and the Belgian as " ? suberectus x Sjirengelii." Of the latter I have seen only dried specimens, which in this instance are not sufficient for distinction, as Dr. Focke points out. Mr. Ley took Dr. Focke and me in 1894 to see his plant on the Beacon Hill, near Trelleck, Monmouthshire, where it "occupies a large area of woodland (some three square miles)," and his article describing the plant (Journ, Bot. 1896, 159, 160) contains the following remarks bearing on the question of a possible hybrid origin: "On the adjoining heath occurs what seems to be a form of the same bramble with leaves much more deeply cut and plicate, and with the glands of the panicle-rachis fewer and subsessile. A hybrid also occurs on the heath between the last-named plant and (probably) B. Sprengelii W." Some of the specimens in my herbarium re- present this hybrid, and others varying examples of B. orthoclados — all tending to conlarm Dr. Focke's present view. In our future lists, therefore, it may perhaps be sufficient to give Ley's plant as a hybrid — Sprengelii x 1 plicatus. I have seen specimens of very similar plants from several of our western counties, from Cheshire to W. Gloucester. The B. orthocladus Boul. Eonc. Vosg. 127 (1869) is now " suspected " by Dr. Focke to be B. foliosus x macropliyllus ; but apparently the same plant had in 1861 been named B. anoplostachys by Mueller, its discoverer, and so could have no right to Boulay's name in 1869 (Journ. Bot. 1905, 201). Next to B. orthoclados in Bicbi Europai comes "i?. riibricolor Blox. in Syme Engl. Bot. ed. 3, iii. p. 180, teste J. E. Griffith," without further description than the following : " Frutices humiles, inter se variabiles, sed omnes B. nitido similes et crebre vel parcius glandulosi. Flores rosei. V. v. sp. Bei Bethesda in N. Wales." But I believe it to be impossible to separate Bloxam's plant (now for many years shut out of our list) from the very aggregate B. lentiginosus Lees, which is common and variable in the Bethesda neighbourhood. (To be concluded) PROTECTION OF THE CAPE FLORA. The Selborne Magazine for April contains an interesting account of the steps taken for the protection of the Cape Flora, from the pen of Mr. A. Handel Hamer, Vice-President of the Mountain Club of South Africa. The gradual destruction of the more attractive plants of Table Mountain and other habitats near centres of population had proceeded for some time. At last " public opinion grew strong on the matter, and in 1905 the Cape Government passed an Act under which regulations could be ssued forbidding the sale of certain species, and instituting close- PROTECTION OF THE CAPE FLORA 183 times during which certain species might not be plucked. This legislation, however, had very little, if any, effect, owing to the fact that its administration was left to the police. In 1910 the Mountain Club of South Africa began to take up the matter afresh, and for this purpose one of the effects of the establishment of the Union Government of South Africa was fortunate, inasmuch as the protection of the fauna and flora was given over to the Provincial Councils, which were established for each Province, as the late separate Colonies now became. An influential depu- tation was arranged, and waited upon the Administrator of the Cape Province (Sir Frederic de Waal) to urge the necessity of further steps being taken. The representations of this deputation were well received by Sir Frederic, who has thrown himself actively into the work of constructing the necessary Provincial legislation in all directions. As a result, about twelve months afterwards, new regulations were got out, greatly enlarging the scope of the former ones, and absolutely forbidding the plucking of a large number of species by anybody for a period of three years. At the same time a Wild Flowers Protection Committee was formed by the Mountain Club, who invited delegates from the National Society, the Eoyal Society of South Africa, the South Africa Museum, the South African College, The Publicity Association, and the City Council. From this large committee a vigilance committee was picked of members who were determined to see the regulations administered. " The immediate result was to demonstrate the absolute use- lessness of the Cape Government's Act under which the new regulations were issued. As soon as prosecutions were instituted it was found that they could rarely be successful, because it was necessary to prove that the flowers had been plucked on Crown lands, and there was no means of compelling the accused to say where he had got the flower from. Moreover, according to the preamble of this Act, the species was required to be in danger of extinction." A new ordinance was therefore obtained, of which the practical legislative points are as follows : — " The onus of proof is placed upon the accused that he obtained a prohibited flower in a legitimate manner. Such a legitimate possession could only occur if he were the owner or had the written per- mission of the owner of the land from which it had been obtained, or if it had been cultivated in land set apart for the cultivation of flowers. The selling of blooms, or any other parts of specified plants, by anybody whatsoever (owner of the land on which they grew not excepted) can be prohibited by the regulations, except- ing only flowers cultivated on land set apart for the purpose. Provision is made by which the Commissioner of Police can issue authority to suitable persons, by virtue of which written authori- ties or warrants the holders can require the name and place of abode of any person reasonably suspected of transgressing the regulations. If the required information be refused an additional fine is provided for if the offender be eventually brought to book. These warrant-holders have formed themselves into a special com- 184 THE JOURNAL OF BOTANY mittee, and have drawn up a constitution with a view to keeping themselves up to the mark and providing i7iter alia that they should meet periodically to compare notes, and that no names should be handed to the police for prosecution until the cases had been submitted to the committee or a sub-committee appointed to supervise prosecutions. The reason for this latter provision is, of course, to avoid the possibihty of the movement being injured by hasty or mistaken prosecutions." SHORT NOTES. EosA siNiCA. — Rosa sinica is often cited as of Ait. Hort. Kew. (ed. 2, iii. 261) — e.g. by Lindley, Rosarum Monographia (1820), Hemsley in Index Fl. Sinensis, i. 250, Hook. f. in Fl. Bl-it. Ind. ii. 364 — to designate a species distinct from R. sinica L. Syst. Veg. ed. 13, 394 (1774), a plant with subglobose glabrous receptacle. Alton in his first edition (Hort. Kew. ii. 203 (1789) ) merely copies Linngeus's diagnosis ; the reference in the second edition is a repetition of the first. R. sinica Ait. is therefore identical with R. sinica L. Alton adds " cult. 1759 by Mr. Philip Miller." Miller makes no reference to this species in his Dictionary, and there is no specimen from him in the Banksian Herbarium. Eobert Brown in his MSS. writes: "said to be cultivated in 1759 by Mr. Philip Miller, and as there is no reference to his Dictionary, it must be inserted from the memory of Mr. Alton, the elder." R. sinica as a synonym or closely allied species of R. lavigata starts with Lindley, whose description is based on a specimen from Bladh in Herb. Bank, whose figure is copied from a Chinese drawing in the same collection. R. sinica L. is probably the same as R. inclica L., see Lindley, op. cit. — W. Fawcett. Lecanora isidioides Nyl. in the New Forest. — This extremely rare and local lichen, which has hitherto been recorded in Great Britain from a small area in North Wales only, was found quite recently (April, 1914), growing in a part of the New Forest, near the hamlet of Cadnam. It is on the mossy trunk of an oak tree, southern aspect, and extends in more or less isolated patches from one to eight feet above the ground. The plant is in a very healthy and vigorous condition, both in the development of the thallus and in the production of apothecia. Crombie says [Monograph of Lichens foimd in Great Britain, parti., 1894): — " The thallus is rather scattered, greenish grey when moistened, usually but sparingly fertile, though in one corticolous fragment the apothecia are somewhat numerous." The New Forest specimens are abundantly fertile, so much so that the apothecia are in some cases slightly angular owing to close contact. The fruits are also, on the average, slightly larger than those of the Salwey and Borrer specimens in the herbarium of the British Museum, but this may possibly be due to the age of the specimens. The diameter of the larger apothecia of the New Forest specimens SHORT NOTES 185 is just over 2 ixim. The spores are all slightly constricted, and the mature ones are decidedly brown ; they vary much in size, the largest being 0'03 mm. long and 0'015 mm. broad. The Salwey specimens are dated Cwm Buchen, 1835, and one of Borrer's, 1841. Crombie remarks, in the monograph already referred to, that Lecanora isidioides has not recently been met with, and there is no record from anywhere in Great Britain since the date of the book. — Robert Paulsen. PucciNiA Smyrnii. — This fungus, parasitic on Smyrnium Olusatrimi, is recorded by Plowright as occurring, "^cidiospores, May to June; teleutospores, June to July," and by Grove, in his book, as " ^cidia, April-June ; teleutospores, June-August." It is perhaps, therefore, worthy of record to state that through the kindness of Miss D. E. Gepp I have received consignments of the fungus from Torquay each month from August to April. In every case aecidia and teleutosori were present, the former always much the more abundant. In December, January, and February the teleutosori were very few in number and indifferently hypophyllous or epiphyllous. The teleutospores germinate overnight in hanging- drop, giving rise to a typical promycelium, the sporidia produced often having germinated in situ by the morning. — J. Ramsbottom. Isle of Man Hepatic^ (p. 45). —Prof. Farmer states that " little or nothing has yet been done towards recording the species that occur in the Island." As an official of the Moss Exchange Club I should like to call attention to the second edition of the Census Catalogue of British Heijaiicce, issued in May, 1913, where all the species mentioned by Prof. Farmer are recorded, except Eucalyx suhelUpticus (hitherto only noted for Mid-Perth), Lejeunia cavifolia var. planiuscula and Anthoceros IcBvis. The list given in the Catalogiie is a rich one, based on work done by Mr. G. A. Holt, with supplementary records supplied by Mr. Beesley and others. I understand that Mr. Hunter has recently made further additions of interest. Alto- gether about eighty species and varieties are known to occur in the Island. Of the genera referred to by Prof. Farmer there are recorded three species of Lepidozia, seven species of LopJiozia, and one species each of Fossomhronia, Madotheca, and Radula. — J. A. Wheldon. BE VIEWS. Floioering Plants of the Biviera : a Descriptive Account of 1800 of the more Interesting Species. By H. Stuart Thompson, F.L.S. With an Introduction on Riviera Vegetation by A. G. Tansley, M.A. 24 coloured plates (112 figures), after water-colour drawings, by Clarence Bicknell, and reproductions of 16 photographs of vegetation by the author. Pp. xxviii, 249. 8vo, cloth. 10s. M. net. Longmans, Green and Co. There was certainly room for a portable book descriptive of the numerous flowering plants to be found on the French and 186 THE JOURNAL OF BOTANY Italian Eiviera, and Mr. Stuart Thompson has succeeded in supplying it. For this he is especially well quaHfied, for besides a good general acquaintance with European plants, he has acquired a special knowledge of the plants of the region, more particularly of the Department of the Var, which has an especially rich flora — 2140 species of phanerogams alone being recorded for it. His previous volumes on alpine and sub- alpine plants have moreover already shown his competence for the task. The book is wisely planned so as to include within its covers all that is needed to render the descriptions intelligible to the intelligent amateur. There is a chapter on collecting and presei'ving plants, a glossary of botanical terms, and a synopsis of families. Mr. Tansley's general introductory essay, although comprised within nine pages, is a clear and readable account of the general features of the vegetation, which are also presented by the excellent reproductions of Mr. Thompson's photographs. Keys to the genera are given under each order ; the descriptions of the species should present no difficulties to those who are accustomed to work with a British flora, and indeed can without difficulty be mastered by those who attempt the w^ork of dis- crimination for the first time. They will be aided in their work by the coloured figures, much reduced from the admirable drawings of Mr. Clarence Bicknell, who has generously allowed the author to make use of these. Their insertion may be justified on the principle that half a loaf, or even a smaller portion, is better than no bread ; but those who know Mr. Bicknell's volume, published in 1885, on the Floioering Plants and Ferns of the Eiviera, will share with us the hope that more of his drawings may be reproduced in a style worthy of the originals. Mr. Thompson in his preface usefully summarizes the literature which has already appeared on the plants of the region ; and acknowledges the help he has received from various botanists — a curious sentence towards the end of the penultimate paragraph needs revision. " The nomenclature does not follow rigidly the Vienna Eules of 1905 ; and in some cases a well-known name is purposely left, even though it may not be the earliest." We are glad that Mr. Thompson has not felt it necessary to coin what are called " English names." It remains to be said that the book is admirably printed on thin but opaque paper, is suitably bound and is light in the hand. We anticipate for it a large sale. Genera of British Plants, arranged according to Engler's ' Sijllahm der Pflanzenfamilien' {Seventh edition, 1912), luith the addition of the Characters of the Genera. By Humphrey G. Carter, M.B., Ch.B. Pp. xviii. -f 121. Price 4s. net. Cambridge : at the University Press. Arranged according to any reasonable system of classifica- tion, a book with such a subject should prove a welcome addition to the literature of British botany. So much attention is paid nowadays — too often vainly, it must be said— to hair's-breadth TRANSACTIONS OP THE BRITISH MYCOLOGICAL SOCIETY 187 distinctions between " sub-species," " varieties," " sub-varieties," "forms," and so forth, that this compact Httle vohime dealing w^ith the British flora from a wider outlook comes as a wholesome change. It should be not only a help to the beginner who does not know his genera, but a healthy reminder to the expert species- monger who forgets their existence. The merits of Engler's system appeal most strongly, perhaps, to the systematist whose practical experience has been confined to the flora of the north temperate regions in the eastern hemi- sphere. The general practical systematist, whose daily business it is to deal with tropical plants as well, embracing every genus of every natural order, may regard the older British system of Bentham and Hooker with more favourable eye. Nevertheless, if Engler's system is all that Mr. Carter claims it to be in his preface, he has increased the debt that British botanists will owe him for his lucid exposition of the genera native in their country, by referring them to " the nearest approach to a natural system that we possess." The modification of the nomenclature in Pteridophyta, made to " secure uniformity in the terminations of the names of orders," might have been extended beyond that group, or omitted altogether. Bower's classification of the Ferns, and Warming's arrange- ment of the Urticales, are the only departures from Engler's system adopted ; and these seem desirable in both cases. The book is well printed, and the type differentiated with judgment. The size is not inconvenient, but the volume might have been more pocketable, even at the cost of a little greater thickness. The price, too, is rather high, even for an introduction to Engler's system, when Hayward's Pocket-book can be bought for another sixpence. jj_ p_ Wernham. Transactions of the British Mycological Society for 1913. (Vol. iv. part 2 ; pubhshed May 28, 1914.) Worcester: Baylis & Son. Price 10s. 6d. The increased activity amongst mycologists in this country is indicated by the Transactions of the British Mycological Society for the season 1913. The number before us contains tw^o hundred and twenty-seven pages, and is greater in size than either of the first two volumes, which each occupied five years. There is also a great increase in the number of active members, and there can be no doubt that the Society is flourishing in every way, and fulfilling all the functions that such a society can. The informal spring foray held at Dolgelley is reported, and a list of the fungi and mycetozoa found there is given. The autumn foray at Haslemere is described at length, and the fungi found there are listed ; Mr. Carleton Rea is responsible for these accounts. The mycetozoa of the Haslemere foray are recorded by Miss G. Lister, and short notes are added in certain cases. The President for the year, Mr. A. D. Cotton, gave as his address " Some Sugges- tions as to the Study and Critical Revision of certain Genera of the Agaricaceje," in which he urged that the monograph was the 188 THE JOURNAL OF BOTANY means by which any real advance would be made, and pointed out various recent ideas on intensive description. Mr. Cotton also has a short paper on the production of impoverished spores by decapitated agarics. Other papers on Basidiomycetes are : — " The Fruit-body Mechanism of Bolbitius," by Professor A. H. R. Buller, in which it is shown that this genus stands far apart from Coprinus ; " Some Notes on the Genera of the Thelephorese," showing the origin of the present ideas on the group, by Miss E. M. Wakefield, who also has a note " On the Identity of Corti- ciuvi porosum Berk, et Curt," which is held to be identical with the C. stramineum of Bresadola. Mr. F. T. Brooks records his observations on pure cultures of certain Ascomycetes and Basidiomycetes, the lignicolous species dealt with being grown on sterilized blocks of wood. In the Uredineae, Mr. J. Eams- bottom deals with the nomenclature of some of the species according to the International Rules of Nomenclature. Mr. W. Watson describes as new a Pyrenomycete, Pleospora hepaticola. In the Discomycetes, Mr. Ramsbottom gives " Some Notes on the History of the Classification of the Discomycetes," beginning with Theophrastus and ending with the present systems ; and " A List of the British species of Discomycetes arranged accord- ing to Boudier's System, with a Key to the Genera." The names given in Massee's " Fungus Flora " are appended. Dr. Bayliss Elliott describes, diagnoses, and illustrates a new variety of mould, Sepedonium mucorinum Harz var. botryoides. New records for this country and new species to science are dealt with by Dr. J. W. Elhs (" New British Fungi "), Miss A. Lorrain Smith and Mr. J. Ramsbottom ("New or Rare Microfungi"), and Mr. Rea (" New and Rare British Fungi "). The last paper is illustrated by three plates, two of which are coloured, and are from drawings by Mrs. Rea. The results obtained by cytologists in the study of the reproductive processes of fungi during the past year are dealt with by Mr. Ramsbottom. It will be seen that all branches of the study receive attention in the Transactions, and both the Society and their indefatigable secretary, Mr. Carleton Rea, are to be congratulated on the appearance of a handsome number. Die Stoffivanderung in ahleheyidend Blattern. Von Dr. Nicolas Swart. Pp. 1-118, 5 Tafeln. 6 Mark. Jena: Gustav Fischer. This interesting addition to the literature of vegetable physiology begins with a general introduction of a couple of pages, occupied chiefly with a discussion of previous conclusions, such as those of Ebermayer (1876) concerning the behaviour of various food-materials stored in the leaf in the autumn up to the time of its fall. The opinion of the earlier students of this sub- ject was that the food-materials in question travelled into the stem, or perennating parts, when the deciduous leaves made their preparations for fall. Wehmer's subsequent work (1892) and criticism of this is mentioned, especially in regard to erroneous deductions made from ash-analysis. The rest of the work is divided into three parts. The first (pp. 4-69) deals with the UNTERSUCHUNGEN UBER DIE FLECHTENGONIDIEN 189 " Auswanderungslehre " historically, affording a valuable risumi of previous researches from Sachs (1863) onwards, and con- cluding with tabular details of the author's ash-analyses of various plants of widely-differing affinities ; the respective weights of the various food-materials in the green leaf, and the yellowing, falling leaf are compared. Part ii. (pp. 70-96) deals with the colour- change induced in the leaf before it falls — namely, from green to yellow. The subject is viewed from many points : the arrange- ment of the chlorophyll ; anatomical changes in the leaf-base ; microscopic changes in the chloroplasts ; climatic influences ; effect of anthocyan upon translocation of food-stuffs. This part concludes with the statement that the yellowing of the leaf is not the result of its gradual dying, but a vital process, the visible effect of various physiological changes. Part iii. (pp. 97-117) is entitled " Schlussbetrachtungen," and deals with the general causes of leaf-fall in evei'green plants and in deciduous trees. Finally, the causes of the loss of food-material from leaves before their fall is discussed. This loss, it would seem, is not due to such a simple process as the mere travelling of the substances into the permanent parts, but is an essential portion of the com- plicated changes, structural and chemical, inseparable from the phenomenon of leaf-fall. The subject before us is one aspect of the wider familiar proposition that leaf-fall is essentially a pro- cess concerned with life, not death. And this process is so com- plex, the changes presumably so continuous and covering so long a period, that the observations of a single student, taken even with that scrupulous care of which Dr. Swart's work bears the unmistakable impress, must needs be somewhat unconvincing, if only for the practical limits placed upon their frequency during that period. This book of his is, nevertheless, of considerable historical interest, and is written with a lucidity and conciseness not always common in the work of his countrymen. The author's own results are not very considerable, but they should afford a useful indication of the general lines upon which co-operative research might profitably be directed. H. F. Wernham. Unters2ichungen iiber die Flechtengonidien. Von Fredr. Elfving. Acta Soc. Sci. Fenn. Helsingfors, 1913. Tom. xliv. No. 2. 71 pp. ; 8 plates. In this publication Herr Elfving has revived the old con- troversy as to the origin of the green cells or gonidia in the lichen. The dual nature of the thallus has been so long accepted, and has fitted in so exactly with the life conditions of the conjoint organ- isms, that it gives one a considerable shock to be taken back to the position held by Tulasne, and to find it again seriously main- tained that the gonidia are genetically connected with the hyphae. The author has not tested his theory by cultures — which alone would be decisive— but by examination of the growing areas of the thallus. He claims to have seen the different stages of forma- tion of the gonidia in a number of lichens associated with such 190 THE JOURNAL OF BOTANY different " algae " as Cystococais, Trentepohlia, Stigonema, and Nostoc. He allows that those gonidia increase by division within the thallus after they have been formed by the hyphae, and that they also may live a free life in the open as " algae." Elfving's view lands us in a series of problems : how are we to account for the origin of other algae that do not enter into symbiosis with the lichen fungus, but the life-histories of which are entirely comparable with that of gonidia in a free condition, unless, as he seems to hint, the ancestors of those algae are to be found among lichen gonidia ? Again, how explain the twofold repro- ductive system combined in one plant — the fungal and the algal ; each after its kind '? It is easier to suppose wrong observations as to the genetic connection of the colourless and the coloured cells, than to accept the conclusion that a homogeneous plant should exactly follow the life-history of two different groups of plants, and of various sections within that group. The author does not attempt to explain these anomalies ; he is content to affirm and reaffirm the correctness of his own observations. a t o A. Li. b. BOOK-NOTES, NEWS, dc It has required large type, wide leading, and broad margins to extend to less than a hundred small pages Dr. William Macdonald's reprinted papers on Makers of Modern Agriculture (Macmillan). The "makers" are Jethro Tull (1674-1740), Thomas William Coke—" Coke of Norfolk "—(1752-1842), Arthur Young (1741- 1820), Sir John Sinclair (1754-1835) and Cyrus H. McCor- mick (1809-1884) — the last a Virginian, " the inventor of the reaper" — i.e. the reaping machine. The sketches betray evi- dence throughout of their newspaper origin ; thus the notice of McCormick begins : — " It is hardly to be expected that those people who devoutly chant in a million churches the fourth sentence of the Lord's Prayer should think with gratitude of any other person than the Divine Giver of all Good. Yet it is strange to reflect that, although every schoolboy knows something of the life of our least Poet Laureate, not one in ten thousand could tell you the career of the man who responded in a truly miraculous manner to the heartfelt, world-voiced matin of both rich and poor, ' Give us this day our daily bi'ead.' " The knowledge which Macaulay ascribed to his schoolboy pales before that which Dr. Macdonald postulates for his. The " subsequent paper," promised on p. 52, does not appear : it was to deal with the writings of Young, of which Dr. Macdonald says : " Our library is far from complete, yet we possess sixty-six volumes of his sparkling prose, which placed one upon another attain a height of nine feet — a monu- ment of amazing industry .... He met and conversed with the greatest savants of the age, yet his mind never burst the old wine bottles which he served out in the Sussex store." " Spai'kling prose" of this kind adorns the little book throughout, but we cannot help thinking it is dear at half-a-crown net. BOOK-NOTES, NEWS, ETC. 191 The Country Month by Month, in which Mrs. Owen (better known as the editor of " A Son of the Marshes ") and Mr. G. S. Boulger combined " to try and give a practical direction to lovers of Nature in their observations by teUing them of the sights they may expect to find, month by month, in their country wanderings," has now been reissued in a handsome voUime of 500 pages, whicli is rendered attractive by the addition of twelve coloured plates and twenty illustrations from photographs. The names of the authors are sufficient guarantee that the slipshoddity which still characterizes too many "popular" works is absent from this, and we know of no better book for the dweller in the country who wants to know something about the birds, plants and insects — the first dealt with by Mrs. Owen, the others by Mr. Boulger — which he may meet with on his walks. The book, which costs 6s. net, is published by Messrs. Duckworth, who still follow the practice, which we hoped had become obsolete, of dis- figuring the titlepages of their " review " copies with an ugly rubber stamp in violet ink. At the meeting of the Linnean Society on June 18th, a paper by the late Mr. William West, "Ecological Notes, chiefly Crypto- gamic," was read in abstract by Dr. Stapf, who remarked that this paper was the outcome of a suggestion by Prof. Engler, that whilst abundance of observations existed of ecological facts regarding phanerogams, the cryptogams had been neglected. This paper, intended as the first of a series, deals chiefly with the corticolous associations of epiphytes. A very large number of observations were detailed, and an approximate percentage estimate of the chief epiphytes was given. Stereodon cupressi- formis var. filiformis is found to be the most prevalent epiphytic moss generally ; but in some localities, especially those with a very heavy rainfall, Isothecium myosuroides becomes the most abundant, especially on the lower part of the trunks. Parmelia saxatilis is the most abundant epiphytic lichen : P. physodes sometimes becomes dominant in exposed and wind-swept places. Lecanora tartarea and Platysma glaucum attain dominance some- times in places subject to frequent montane storms. Fndlania dilatata is probably the most frequent epiphyte among the Hepatics, but Metzgeria furcata is now and then the most abun- dant. From the detailed tables given of epiphytes, which only represent portions of the district examined, and are not given as what may be expected in every locality, the percentage comes out: — Stereodon cupressiforniis 16, Parmelia saxatilis 6, Isothe- cium myosuroides 2, Frullania dilatata 2, Parmelia fuliginosa vsiv. Icetevirens 2, Lecanora tartarea 2, Platysma glaucum 1, and various species of Pertusaria 1. The observations extend over parts of Scotland, Wales, Ireland, and the Lake District. The following British botanists will attend the forthcoming meeting of the British Association in AustraHa : — Prof. F. 0. Bower, President of the Botanical Section ; Prof. A. C. Seward, Vice-President ; Misses M. Benson, E. M. Berridge, L. J. Clarke, 192 THE JOURNAL OP BOTANY L. S. Gibbs, E. E. Saunders, A. L. Smith, and E. N. Thomas, and Messrs. W. Bateson, E. H. Compton, G. C. Druce, A. H. Evans, H. 0. Forbes, E. P. Gregory, W. P. Hiern, T. Johnson, and Dr. A. B. Eendle. In a paper entitled Beitrdge zur Kenntnis der Pteridophyten- gattung PhyUitis (reprinted from Q^sterreich. Botan. Zeitschrift. 1914, pp. 19-36, 5 pis. and 2 maps), Friedrich Morton publishes two studies of the genus PhyUitis Ludw. [Scolopcndrinm Adans.) : (i.) the finding of P. Heviionitis (Lag.) 0. Kuntze in the Quarnero region, and the distribution of the species ; (ii.) the systematic position, distribution, and ecology of P. hybrida (Milde) Christensen. Having made a comparative study of this species and its allies in respect of the following structures : (1) the endings of the vascular bundles in the frond-lobes ; (2) the posi- tion of the sori ; (3) the indusium ; (4) the layers of tissue from which the indusium arises, he is of opinion that P. hybrida is no hybrid, but an independent species, occupying a systematic posi- tion between Ceterach and P. Scolopendrimn, similar to that of P. Hemionitis, to which, in its anatomy and morphology, it is most nearly akin. It is endemic in the southern Quarnero Archipelago, in the Adriatic Sea. We regret to note the death, at the age of seventy-five, of M. Philippe van Tieghem, Professor of Botany at the Paris Natural History Museum. Van Tieghem's most important botanical work was on the comparative anatomy of the vegetative structure of the flower in numerous families of seed-plants. Many of these communications were published in the Annales des Sciences Naturelles, of which he was botanical editor from 1882, when he succeeded Decaisne, until the time of his death. One of the best known and most important appeared in 1869 on the comparative anatomy of the female flower and fruit in Gymnosperms, in which he attacked the problem of the morphology of the cone-scales from an anatomical point of view, and demonstrated a uniform plan of structure in this organ throughout the Pinacete. In 1884 he produced his Traiti de Botanique, the second edition of which (1891) contained 1855 pages. His systems of classification were, as far as concerned the flowering plants, based primarily on characters of the ovule ; they were conceived on too narrow lines, besides being bur- dened with a novel and extensive terminology. Van Tieghem also made valuable contributions to the study of the phycomycetous fungi , especially on the morphology and physiology of the Mucorineas. The Journal of the Royal Horticultural Society (xxxix. part 3, April) contains an interesting illustrated paper by Mr. E. A. Bunyard on " The History and Development of the Strawberry," and a lecture by the Eev. G. Henslow on " The Evolution of Plants and the Directivity of Life, as shown by the Eeproductive Organs " ; as well as contributions from the Wisley Laboratory and many papers dealing with horticultural matters. The degree of D.Sc. in Botany has been conferred by the University of London on Mr. H. F. Wernham, of the Department of Botany, British Museum. Journ.Bot. Tab. 532. P.Highley del.et lith . West.Kewman imp. Poa remctiflora Murb. /"exilis. 193 POA EEMOTIFLOEA Murb. IN JEKSEY. By C. E. Salmon, F.L.S. (Plate 532.) In the Journal of Botany for 1913, p. 16, mention was made of a Poa found in the Channel Islands in 1910, and thought to be new to Great Britain and the Sarnian flora. This diminutive grass seems to have lieen first observed by Tommasini in 1872, and to have been first published by J. Freyn in his " Flora von Siid-Istrien " (Zool.-Bot. Gesellsch. Wien, xxvii. 469 (1877) ) as "Poa annua L. /3 exilis Tommas. ined., 1-3 stemmed, in minute rosettes, only 2-8 cm. high ; panicle few-flowered, spike-like or with only 1-2 horizontally-spreading branches at the base . . . (Tommasini, 1872)." In 1895 Battandier and Trabut (Fl. de I'Alg^rie, 206) introduce the grass as P. annua L. /5^ remotiflora Hackel, with the follow- ing meagre description : " Epillets allonges lin^aires, 5-6 fleurs espac6es." It was, however, left to S. Murbeck to work out minutely this interesting grass, and the results of his careful examination may be found in his Contribution de la Flore du Nord-ouest de I'Afriqne, iv. 22 (1900), where he compares it with allied forms, and gives valuable figures. I cannot do better than quote his description verbatim, omitting as unnecessary for British botanists the contrasting characters of P. dimorphantha, " P. remotiflora (Hack.) nob. — Nova spec. — Annua, foliorum fasciculis sterilibus destituta. Culmi plerumque numerosi, fasci- culati, e basi arcuata adscendentes, 5-20 cm. alti, Isevissimi. Vaginae glaberrimae, Iteves, inferiores compressae et dorso carinatae. Ligula foliorum inferiorum truncate-, superiorum triangulari- oblonga. Laminae lineares, planae vel subcomplicatae (explanataa l'5-4-5 mm. latae), laete virides, moUes, margine et ad nervum medianum scabriusculae, caeterum glaberrimae, etiam foliorum inferiorum apicem versus sat subito acutatae. Panicula triangulari- ovalis vel oblongo-rhomboidea, duplo vel triplo longior quam latior (longit. 3'5-7'5 cm., latit. 1-5-3-5 cm.), unilateralis, laxa, ramis 2-3-nis, laevibus, erecto-patulis, patentibus vel infimis post anthesin patentissimis attamen non refractis. Spiculae oblongae vel oblongo- lineares, acutiusculae, a lateribus compressse, 2-5-florae, 3"5-5-5 mm. longae, virides vel violaceo-variegatae, rachilla glaberrima. Gluma inferior ovato-oblonga, subacuta, 1-nervis, 1-1-3 mm. longa ; superior oblonga vel ovato-oblonga, obtusa vel acutiuscula, 3-nervis, 1-5-2 mm. longa. Flores in quaque spicula remoti, rachillam non vel vix occultantes, sat dissimiles: zw/c?-iores hermaphroditi, sub- proterandri, oblongo-lanceolati, 2-2-8 mm. longi; glumella inferior dorso rotundata, a basi usque ad ultra medium nervis 5 dense sericeo-villosis ornata, apice membranaceo rotundato-obtusa ; glumella superior inferiorem subaequans, a basi fere usque ad apicem ad nervos sericeo-villosa ; lodiculae oblique triangulares ; Journal of Botany. — Vol. 52. [August, 1914.] q 194 THE JOURNAL OF BOTANY filamenta staminum glumellis non vel vix longiora; antherae 0-22-0-33 mm. longge, albidae, loculis ovoideis, circ. duple longi- oribus quam latioribus ; caryopsis oblongo-ovoidea, quam glumella inferior circ. tertia parte brevior ; flos supremus nunc herm- aphroditus nunc femineus, post anthesin ovoideo-lanceolatus, acutiusculis, 1-8-2 mm. longus, stipiti eo tertia vel quarta parte breviori insidens, erectus ; glumella inferior superiore paullo longior, ut superior fere usque ad apicem ad nervos sericeo-villosa ; lodiculae minutissimae, triangulares ; caryopsis ovoidea, quam glumella inferior tertia vel quarta parte brevior. — Fl. & fr. Mart. " Syn. P. annua var. remotiflora Hack, in Baenitz Herb. Europ. Lief, xxxix. (1880), n. 3999 (sine diagn.). P. annua fi remotiflora Hack, in Batt. & Trab. Fl. Alg6r. v. 206 (1895). "Icon. Tab. nostra, xiv. fig. 12. ''Distribution geographique. Algerie: Lieux humides entre Philippe-ville et Damr6mont (Murb. & Olin.) ; d'apr^s Batt. & Trab., I.e., aussi a Kouiba et a Teniet-el-Haad. — Grece : In locis incultis prope Athenas (Heldreich, 1878). " Cette plante, distinguee il y a d^ja une vingtaine d'ann^es par Hackel comme vari^te du Poa annua L., mais dont une description n'a et6 publiee qu'en 1895 par Trabut," est tres voisine du P. annua, auquel elle a ete r^unie aussi par cet auteur. Elle ne parait toutefois pas avoir et6 jusqu'ici etudi6e en detail. En reality, elle presente plusieurs caract^res importants, qui la differencient de I'espfece de Linne (voir ci-dessous). L'extr^me petitesse des anth^res, qui sont constamment 2 a 3 fois plus courtes que dans cette derniere, me parait surtout constituer une difference si essentielle, qu'il est pr6ferable de regarder cette plante comme une esp^ce a part. " Un Poa remotiflora ['Poa {ArctopJiila) remotiflora'] se trouve d6ja dans Kuprecht (Fl. samoj. p. 63, 1845). — Mais comme la plante de Euprecht appartient au genre Colpodium Trin. {Arcto- phila Eupr.) et qu'il n'est pas probable qu'elle constitue une espfece distincte du Colpodium pendulimmi Griseb., je n'ai pas vu d'inconvenient a employer ici comme designation specifique le nom de vari^te donne par Hackel. " On trouvera ci-dessous un aper9u des caract^res par lesquels se distinguent entre eux les Poa dimorphantha, remotiflora et annua. II se fonde, comme les descriptions donuees plus haut, non settlement sur des specimens s6ch6s, mais aussi sur des ex- emplaires vivants, cultiv^s dans le Jardin botanique de Lund. "P. remotiflora. " Cliaumes arqu6s-ascendants. "P. anmia. " Chaumes ascendants ou couches. " Glumes inferieures com- prim6es, car6n6es sur le dos. " Glumes infirieures sub- cylindriques, b> peine carenees sur le dos. • Trabut la decrit ainsi (I. c.) : " Epillets allonges lin^aires, 5-6 fleura espac^es." POA REMOTIFLORA IN JERSEY 195 "P. remotiflora. ''Feuilles scabres sur les bords ; meme les inf^rieures assez brusquement attenu6es vers le so m met. " Panic ide ovale-triangulaire ou oblongue-rhomboidale, de 1-5 a 3 fois aussi longue que large, semi-conique (a base semi- orbiculaire) ; rameaux infer. 6tal6s apr^s I'anth^se. " Fleurs espac6es, assez dis- semblables, les infer ieures (de 1^4) longues de 2 a 2-8 mm., hermaphrodites, lanceol6es-ob- longues apr6s I'anth^se ; les superieiires (1 ou 2) gen6rale- ment femelles, subaigues, 6troitement ovoides apr6s I'an- these, longues de 1'8 a 2 mm., la terminale de^k^ plus longue que son p^dicelle. " Filets des 6tamines pas ou a peine plus longs que la glu- melle inferieure et n'atteignant pas meme le milieu de la fleur superieure contigue. " Antheres longues de 0-22 k 0-33 mm., a loges ovoides, environ 2 fois aussi longues que larges." " P. annua. " Feuilles scabres sur les bords ; meme les inf^rieures brusquement contract^es au sommet. " Panicule triangulaire, de 1*2 a 1-6 fois aussi longue que large, semi-conique (a base semi- orbiculaire) ; rameaux etales ou r6fi6chis apr6s I'anth^se. " Fleurs imbriquees, peu dis- semblables, les infirieures (de 1 a 5) longues de 2-5 a 4 mm., hermaphrodites, lanceol^es- ou ovales-oblongues apr6s I'an- these ; les sup&rieures (1 ou 2) g6n6ralement femelles, aigues, 6troitement ovoides apr^s I'an- these, longues de 2 a 2-5 mm., la terminale au moins 2 fois aussi longue que son p6dicelle. "Filets des 6tamines pas plus longs que la glumelle in- ferieure et n'atteignant jamais le sommet de la fleur sup6rieure contigue. " AntMres longues de 0*6 a 0-8 mm., a loges oblongues- lineaires et de 4 a 5 fois aussi longues que larges." Ascherson & Graebner (Syn. Mittel-eur. Fl. ii. 389, 1900) refer to our plant as P. annua L. subsp. exilis, and consider that the small plant mentioned by Freyn in 1877 should be separated as a form, " fi. Tommasinii. Low-growing, 2-8 cm. high ; panicle few-spiked, clustered together above." They also remark that this form of the subspecies (corresponding to P. annua D. pauci- flora) occurs in their Central European region only, and not the typical state. This is certainly also the Channel Island form. Kouy (Fl. Fr. xiv. 268, 1913) places the plant as a "race" under Poa annua ; he does not differentiate the dwarf state, and mentions localities in Corsica and the Department of Var. The chief distinctive characters of P. remotiflora may be easily contrasted with those of P. annua in the table given by Murbeck, the more obvious features being the oblong spike (often quite simple in f. exilis), the panicle-branches non-reflexed after flower- Q 2 196 THE JOURNAL OF BOTANY ing, the non-imbricate flowers and the very small anthers. These points serve to distinguish the plant from the dwarf, starved pale- yellow states of P. annua of our trodden path-margins and dry places. I am much indebted to Dr. O. Stapf for finally determining the Jersey specimens ; he remarks: "It differs from the typical Mediterranean form in having smaller, relatively stouter florets with longer hairs." These characters will be useful to those who separate the dwarf f. exilis from typical remotiflora. The nomenclature is : — PoA REMOTIFLORA Murb. Coutrib. Fl. Nord-ouest Afrique, iv. 22 (1900). P. annua L. /3 remotiflora Hackel in Battand. & Trabut, Fl. Alg^r. v. 206 (1895). P. amiua L. subsp. exilis Aschers. & Graeb. Syn. Mittel-eur. Fl. ii. 389 (1900) ; Battandier, Supp. Fl. Alger. 90 (1910). P. annua L. "race" P. exilis Eouy, Fl. Fr. xiv. 268 (1913). Forma exilis. P. annua L. fi exilis Tommas. ex Freyn, Fl. Sud-lstrien, Zool.- Bot. Gesellsch. (Wien) xxvii. 469 (1877). P. annua L. subsp. exilis Aschers. & Graeb. forma B. Tom- masinii Aschers. & Graeb. Syn. Mittel-eur. Fl. ii. 390(1900). Distribution. — Jersey : Sandy places near the sea, March, 1877, J. Piquet ! (Hb. Mus. Brit.) (f. exilis). Near Fort Eegent, West Mount, and other places, frequent, J. W. White t£' C. E. S., 1910 (f. exilis). Guernsey : Scart Point, Ap. 1891, I. H. Burkill (Hb. Kew) (f. exilis). France, S. : Var. Portugal. Middle and Southern Italy. Corsica. Sicily. S. Istria. Greece ! Morocco. Algeria. Cyrenaica. Syria. Persia. Exsicc. — Baenitz, Herb. Europ. Lief, xxxix. (1880), no. 3999 ; Heldreich, Herb. Gra^c. Norm. 1098 ! Icon. — Murbeck, 1. c. t. xiv. fig. 12. Explanation of Piate 532. 1-2. P. remotiflora f. exilis, drawn from Jersey specimens, natural size. 3. Ditto, luxuriant. 4. Spikelet, enlarged four times. NOTES ON BRITISH PLANTS. By C. E. Moss. III. The Genus Alsine. Since the appearance in this Journal (xxxvii. 317, 1889) of Hiern's article on Alsine, there has been much uncertainty in the minds of British botanists regarding the names of the plants which are usually placed in the genus. Mr. Hiern, it will be remem- bered, substituted Alsine (Hiern) for Spergidaria, and Minuartia (Hiern) for Alsine (Gaertner). H. &. J. Groves, in their edition of Babington's Manual (1904), adopted both of Hiern's substitutions, although they retained a " suborder Alsinece " for a group of plants THE GENUS ALSINE 197 without an Alsine, and placed their Alsine (Hiern) in a " suborder Pol//ca)'pc(e." Eendle & Britten (List, 1907) also adopted Minic- arlia (Hiern) and Alsine (Hiern). On the other hand, Druce (List, 1908; Hayward's Bot. Pocket Book, 1909), Marshall (Lond. Cat. ed. 10 (1908)) and Carter (Gen. Brit. Plants, 1913) continued the union of Alsine (Gaertn.) with Arenaria, after de Candolle (Prodr. iv. 401, 1824) and Bentham & Hooker (Gen. PI. i. 150, 1862). Syme (Eng. Bot. ii. 107, 1864) and WiUiams (in Journ. Bot. xxxiv. 427, 1896) had both adopted Alsine (Gaertn.). I do not think there is very much to be said in favour of the union of Alsine (Gaertn.) and Arenaria, unless a great many other genera of the family Dianthacece (or Caryophyllacece) are also united. It is admitted that the modern genera of the family are very closely allied, and that they are indeterminable in the absence of ripe fruit ; but it is at least in keeping with the scale all but universally adopted for the other genera of the family to retain one name {Alsine Gaertn. or Minuartia Hiern) for the plants whose capsules dehisce with the same number of teeth as the ovary has stigmas, and another name {Arenaria) for the plants whose capsules dehisce with twice as many teeth as the ovary has stigmas. The union of both groups of plants in one genus would logically demand the union of many other genera of the family ; and, unless the changes were very comprehensive indeed, there would be no gain from the practical point of view of identifying the genera without fruit, whilst the changes of specific names on the large scale that would be necessary to bring about this practical benefit would probably not meet with general acceptance. Hence, the points that have to be decided are : (1) whether Alsine (Gaertn.) shall be retained in the commonly accepted sense, or (2) whether Alsine (Hiern) shall be applied to Delia (Dum.) and the name Minuartia (Hiern) given to Alsine (Gaertn.). I must state at the outset that I am not in sympathy with the changing of any established generic name on grounds of priority alone ; and, in the Cambridge British Flora I do not propose to make a single generic change of this kind. This decision is in the fullest accord with the general aims and spirit of the Inter- national Eules, and indeed is logically demanded from those who accept the principle of a list of nomina conservanda of genera. The changes in the names of species which result from the substitution of one generic name for another are too numerous to allow established names of genera to remain long in the arena of nomenclatorial conflicts. Whenever it is proved that the recog- nized name of any genus is incorrect from the point of view of priority, the already established name of that genus should be automatically placed on the list of nomina conservanda. Only those botanists who have given special attention to generic names realize how many changes would be involved if the so-called "law " of priority were rigidly followed in the matter of the names of genera. In fact, I personally regret there is no list of nomina conservanda of species ; but the task of preparing such a list would be colossal ; and there is no wonder therefore that the Vienna 198 THE JOURNAL OF BOTANY Congress shrank from a Herculean undertaking and contented itself with laying down a rule of priority for the names of species. Keturning to the question of Alsmc (Gaertn.) versus Minu- artia (Hiern), the matter is not a very complicated one if the history of Alsine is chronologically considered. Linnaeus, in the first edition of his Species Plaiitarum (p. 272, 1753), had only two species of Alsine. The first was Alsine media L. and the second /I. segetalis L. It is clear that if priority alone be allowed to determine the issue, the name Alsine should be used for one of these two species. Each species has its claim. Scopoli (Fl. Carn. ed. 2, i. 224, 1772) took the first of the two species, and defined his Alsine to include it and it alone. This species is now universally regarded as a Stellaria — it is the common chickweed, S. media Villars ; and hence the Alsine of Scopoli (1772) sinks in Stellaria L. (1753). Hiern {loc. cit.) therefore fixed his attention on the second species, A. segetalis L. This species is sometimes placed in Spergularia and sometimes placed in a separate but closely allied genus, Delia. Mr. Hiern retained Alsine (Hiern) for Spergularia (including Delia), and gave the requisite new combinations for the British species involved. Spergularia, however, is now a nomen conservandum ; and to those of us who accept the nomina conservanda, Hiern's changes under this heading are now obsolete. If, however, A. segetalis L. is placed in Delia (Dumortier Fl. Belg. 110, 1827) as is done by some authorities (e. g., Ascherson & Graebner Fl. Nordostd. Flachl. 316, 1898), Hiern's view would still hold good — if, as I have said, priority alone be allowed to determine the matter. It is doubtless from this point of view that Schinz & Keller (Fl. Schweiz ed. 3, 204, 1909), whilst retaining Spergularia, adopt Alsine (Hiern emend.) for Delia. This change having being accomplished, it remained to find a name for the plants which have been usually referred to Alsine (Gaertn.). Hiern utilized Minuartia (Hiern) for the purpose, and made the necessary new combinations for the British plants ; and, as I have already stated, Hiern has been followed in this matter by H. & J. Groves and by Eendle & Britten, and also by Schinz & Keller {op. cit. p. 200). The position of these botanists is perfectly logical, if judged from the standpoint of priority. There is, however, another point of view to be considered, as will be seen by continuing the study of the historical development of the Linnean concept of Alsine. In the second edition of the Species Plantarum (p. 389, 1762) Linnseus added a third species to his genus. The added species was Alsine mucronata L., which has been an Alsine ever since in the great majority of books on botany. Linnaeus placed this species between the other two, his species now being respectively : (1) A. media L. { — Stellaria media Villars), (2) A. mucroyiata L., and (3) A. segetalis L. Gaertner (Fruct. ii. 223, t. 129, 1791), in founding his genus Alsine, took the second edition of the Species Plantarum as his starting-point. He rightly passed over the first species {A.mcdialj.), THE GENUS ALSINE 19'J as this had ah'eady been correctly placed in Stellaria by Villars ; and then it was perfectly natural that he should take the second species {A. mucronata L.) as the type of his genus. Thus, from Gaertner's point of view, the third species {A. segetalis L.) does not enter into the discussion. Gaertner's view was adopted by Wahlenberg (Fl. Lapp. 1812; Veg. Helv. 1813; Fl. Suec. 1826), who made the necessary new combinations for the species as he dealt with them in his various works. Wahlenberg has been followed by the vast majority of botanists, including Syme & Williams in this country. Wahlenberg, in fact, is usually cited as the authority of the genus, though Wahlenberg himself refers his Alsine to Gaertner. It is, from my point of view, quite immaterial whether Gaertner or Wahlenberg is cited as the authority : the genus is the same in either case, even though Wahlenberg has referred species of Spcrgularia to it ; it is, of course, erroneous to ascribe it to Scopoli (c/. Dalla Torre & Sarn- theim Gen. Siphonog. 157, 1900), for the Alsme of Scopoli, as has been shown, has nothing to do with Alsiiic of any European botanists and must be sunk in Stellaria L. With regard to Gaertner's use of the second edition of the Species Plantarum, it must be borne in mind that the first edition was a very rare book in his day, and, indeed, almost inaccessible to the great majority of the botanists of the eighteenth and of the first half of the nineteenth centuries. A reprint of the first edition, and greatly increased facihties of travel, have rendered the first edition accessible to the majority of botanists ; but the early botanists adopted the only sensible and practicable course when they used the second rather than the first edition of the Species, because they could not check or verify the names of the first edition. The fashion of citing the second edition as the starting-point still persists, as, e.g., in Eouy & Foucaud's Fl. de France ; and it is one of the easiest tests of a mere copier to find if he adds " 1753 " to the page of the second edition. It is a question whether it would not have been better, in the interests of nomenclatorial stability, for the Vienna Congress to have adopted the second instead of the first edition as the commence- ment of the names of vascular plants, because the second edition had virtually, for the reasons just given, been taken as the starting-point of nomenclature by nearly all the earlier post- Linnean botanists, who have thus impressed the names of the second edition so deeply in botanical literature and botanical thought that it is useless nowadays to attempt to change them."'- * It was with all this in mind that I decided (see Cambr. Brit. Fl. ii. p. xiii. 1914), iu order to avoid some undesirable changes of specific names which would be necessitated by rigidly adhering to the first edition of the Species FlaiUariun as the starting-point of nomenclature, to begin with the second edition in the case of those species which were subdivided into two or more species by Linnaeus himself in the second edition. In the case of these latter species, it is the names of the second edition that have become common to all countries ; and thus this slight departure from the letter of the Vienna Rules will always result in conserving established specific names. I can conceive of no one, except the stickler for mere priority, objecting to this little innovation. 200 THE JOURNAL OP BOTANY Bearing in mind that Alsine (Gaertn., Wahlenberg) is the name that is to-day in ahuost universal use and that it has been so since the time of Wahlenberg, those botanists who accept the principle of a list of nomina conservanda of genera, can offer no legitimate objection against the suggestion now made of placing Alsine (Gaertn., Wahlenberg) on the list in question. It seems to me that this is the only legal method of avoiding the undesirable substitution of the name Minuartia (Hiern) for the established name Alsine (Gaertn., Wahlenberg) : the plan would leave Delia free for those who desire to use it, and seems indeed to be the logical corollary of the conservation of Sjjercjularia. The following citations of the British species are given on the assumption that Alsine (Gaertn., Wahlenberg) is regarded as a nomen conservandum. Alsine L. Sp. PI. ed. 2 (non ed. 1) emend. Gaertner Fruct. ii. 223, t. 129 (1791) ; Wahlenberg Fl. Lapp. 127 (1812) ; Fenzl in Endlicher Gen. PI. 964 (1836-1840) ; Bentham & Hooker Gen. PI. i. 150 (1862) as a section ; Pax in Engler & Prantl Pflanzenfam. iii. lb, 82 (1888) ; Rouy et Foucaud Fl. France iii. 261 (1896) ; non Scopoli nee Hiern nee Schinz & Keller; Minuartia L. Sp. PI. ed. 1, ampl. Hiern in Journ. Bot. xxxvii. 320 (1899). 1. Alsine steicta Wahlenberg Fl. Lapp. 127 (1812); Spcrgula stricta Swartz in Kongl. Vet. Acad. Handl. Stockholm xx. 235 (1799) ; Arenaria uliginosa [Schleicher ex] DC. Fl. Fran9. iv. 786 (1805)!; Alsine uliginosa Syme Eng. Bot. ii. 115 (1864); Mimiartia stricta Hiern in Journ. Bot. xxxvii. 320 (1899) ; non Arenaria stricta Michaux Fl. Bor. Am. i. 274 (1803). The only British station of this plant known to British botanists is in Upper Teesdale ; but by Rouy & Foucaud {op. cit. p. 266) the plant is also recorded for Ireland. 2. Alsine verna BartHng in Bartling Sc Wendland Beitr. ii. 63 (1825) ; Syme Eng. Bot. ii. 109 (1864) ; Arenaria saxatilis Hudson Fl. Angl. 168 (1762) non L., inch Ar. laricifolia non L. ; Ar. verna L. Mant. PI. i. 72 (1767) ; Ar. liniflora Jacquin Fl. Austr. V. 22, t. 445 (right-hand figure) (1778) ? non L. ; Ar.juni- perina Withering Arr. ed. 3, 424 (1796) non L., incl. Ar. laricifolia non L. ; Al. saxatilis Wahlenberg Veg. Helv. 87 (1813); Minuartia verna Hiern in Journ. Bot. xxxvii. 320 (1899). The British forms of this species are very variable ; and some of them have been given names by British botanists. However, I am not able to state whether or not these names have been correctly applied. 3. Alsine rubella Wahlenberg Fl. Lapp. 128, t. 6 (1812) ; Syme Eng. Bot. ii. Ill (1864) ; Arenaria gieseki Hornemann in Fl. Dan. fasc. xxvi. 5, t. 1518 (1818) ; Ar. hirta [Wormskiold ex] Hornemann o^j. cit. fasc. xxviii. t. 1646 (1823) ; Ar. rubella Smith Eng. Bot. iv. 267 (1824) ; Minuartia rubella Hiern op. cit. 320 (1899). This is an Arctic species not known to occur in Central Europe. THE GENUS ALSINE 201 4. Alsine tenuifolia Crantz ''■'■ Inst. ii. 407 (1766) ; Wahlen- berg Veg. Helv. 86 (1813); Syme Eng. Bot. ii. 112 (1864); Arenaria tenuifolia L. Sp. PI. 424 (1753) ! ; Minuartia tenuifolia Hiern in Journ. Bot. xxxvii. 321 (1899) non Nees in litt. ex Martius Hort. Erlang. 44 (1814) ; M. leiAophijlla H. & J. Groves in Babington's Man. ed. 9, 61 (1904). 5. Alsine sedoides Kittel Fl. Deutschl. ed. 2, ii. 997 (1844) non Froelich in litt., ex Koch Syn. 114 (1835) ; Cherleria sedoides L. Sp. PI. 425 (1753) ; Alsine clierleria Grenier et Godron Fl. France i. 253 (1848) ; Eouy et Foucaud Fl. France i. 253 (1848) ; Al. cherleria Petermann Deutschl. Fl. 851 (1849) ; Syme Eng. Bot. ii. 108 (1864) ; Minuartia sedoides Hiern in Journ. Bot. xxxvii. 321 (1899). I take it that the International Eules, being retrospective in their action, necessitate the adoption of A. sedoides Kittel. This is an Alpine species unknown in northern Europe : there are very few members of the British flora which have such a geographical range as this. 6. x\lsine I'EPLOiDES Crautz Inst. ii. 406 (1766) ; Wahlen- berg Fl. Suec. i. 282 (1826) ; AreJiaria peploides L. Sp. PL 423 (1753) ! ; Honckenia peploides Ehrhart Beitr. ii. 181 (1788) ; Syme Eng. Bot. ii. 106 (1864) ; Minuartia peplo'ides Hiern in Journ. Bot. xxxvii. 322 (1899). The chief points dealt with in this note regarding Alsine are thus summarised : — (1) In the first edition of the Species Plantar itm (1753) of Linnaeus there are two species of Alsine — A. media L. and A. segetalis L. (2) Scopoli (1772) took the first of these as the type of his Alsine] but as A. media L, is now universally recognised as a Stellaria, the genus Alsine Scop, disappears. (3) Hiern (1899) took the second of the above species as the type of his Alsine. Sometimes this species is placed in (S'^;cr- (jularia, ^which is a nomen conscrvand um ; and thus ^4/si»c Hiern is obsolete. By other authorities the species is placed in Delia, and thus, judging by priority alone, Delia Dum. would become Alsine Hiern emend. Schinz & Keller. (4) In the second edition of the Species Flantarum (i. 1762) Linnteus added a third species of Alsine, A. mucronata L. (5) Gaertner (1799) took this third species as the type of his Alsine, and was followed by Wahlenberg and almost all other botanists. The name Alsine, as thus defined, is consequently established firmly in botanical literature ; and, on this ground, it is here suggested that the name be placed on the list of nomina cojiservanda. * Alsine Crantz Inst. ii. 404 (1766) includes Sagina L., Elatinc L., McehringiaL., Bujfonia L., PolycarpumL., Alsine L. (partim), Arenaria L., and Spergula L., but not Cherleria L. 202 NOTES ON DE. FOCKE'S BUBI EUBOP^I (1914). By the Rev. W. Moyle Rogers, F.L.S. (Concluded from p. 182.) B. hypoleucHS Lefv. & Muell. As this is the earlier of the two names suggested in recent years for our plant (the date of publi- cation being B. hypoleucus L. & M. 1859, B. adscitus Genev. 1860), and as there seems to be no question as to the identity of the two plants, it appears best that we should now retain the earlier name hypoleucus — accepted by us in 1905 (instead of B. micans Gren. & Godr.), after the publication of Dr. Focke's work in Aschers. & Graebn. Fl. Mitt. — rather than again follow him in his final pre- ference for B. adscitus. His reason for such preference in Rubi Europ. is as follows : — " Specie! cognitio ex hac descriptione [i. e. Genev. Mem. Soc. M. et L. viii. p. 88] derivata est ; nomen adsciti igitur praefero." I have not, however, seen authentic specimens of Continental hypoleucus ; and so, personally, I should have also preferred the name adscitus, as our typical and widely distributed British plant agrees admirably with Genevier's specimens and description, as well as with the living examples of his plant that I have seen abundantly in the Channel Islands, Brittany and Normandy. Compare Journ. Bot. 1905, p. 202, and my Hand- book of British Bubi, p. 48 {" B. micans Gren. & Godr."). A brief account of my B. Lettii (Journ. Bot. 1901, p. 381) is followed by the note, "Formam borealem B. adsciti esse e speci- minibus exsiccatis suspicor" — a suggestion which I considered only to reject, when I preferred placing it amongst our Egregii because of the Koehlerian-like armature of all its stronger examples. While obviously recalling B. adscitus in the very greyish tint of the whole plant, B. Lettii seems best placed between B. criniger and B. Gelertii. Dr. Focke confines it to Ireland, and thus far the strong, highly glandular type has not been found elsewhere, though what seems to be a weaker form of the same plant occurs in some of our western counties, especially in Cardigan, Salop, and Hereford. Between his B. adscitus and his B. vestitus he also places a plant of the Plymouth neighbourhood (which I cannot now trace) as " B. adscitus x rusticanus. Planta spectabilis, luxurians, alte scandens, thyrsis longis et inflores- centiis compositis amplis patulis ornata. Singuli frutices in eodem loco variabiles ; occurrunt in plantis magnis interdum folia semper fere ternata cum foliolis subtus virentibus. Sterilis videtur. V. v. duce Archer Briggs." Our reasons for preferring the name B. leucostachys Sm. to the slightly later one B. vestitus Wh. & N. — which is generally adopted on the Continent and is still adhered to by Dr. Focke — are given in full in Journ. Bot. 1905, p. 202, and need not be repeated or added to here. This is placed by him in closer relation to B. hypoleucus {B. adscitus) than in our list. It is when he comes to deal with the Vestiti that his sugges- tion of a hybrid origin for some of the plants becomes more NOTES ON DR. FOCKe's RUBI EUROP^I 203 frequent ; and I regret that I cannot refer fully to this part of his work. But the following paragraph is so remarkable that the readers of the Journal may be glad for me to quote it in full : — " Conjecturae do originehybridogena nonnullorumKuborum Vestitis similium. Buhus adscitus et B. vestitns, simili modo ac Bubi Suberecti, gregem naturalem bene distinctam constituunt. Eepe- riuntur vero species complures nonnullis characteribus ad Vestitos, aliis ad diversos Bubos vergentes. Quaeritur anne tales species ambigentes originis hybridogenic sint, praecipue in aevo diluviano vel pliocasno ortae. Illis temporibus nondum species nobis cog- nitae hodiernae floruerunt, sed formae atavaB, quarum proles mutata et genuina et hybridogena nunc Europam incolit." This is fol- lowed by the heading " Species et prospecies, quarum origo hybridogena e Eubis Vestitis suspicari potest," and a table, from the fourteen lines of which the following are examples : — Series vel " Species atavje (vel recentes) parentes. Species hybridogenae. subseiies. " B. adscitus et B. sulcatus. B. leucandrus. Silvatici. B. vestitns et B. sulcatus. B. macrophyllus. ,, „ B. vulgaris. B. pyramidalis. Vestiti. „ B. egregius. B. vmcronatus. Sevii-Egregii. „ B. ccBsius. B. Balfourianus. Corylifolii." B. gynmostachys Genev. This name, given as a variety of B. leucostachys in my Handbook and in Lond. Cat. ed. x., is now, contrary to Dr. Focke's earlier views (for which see Journ. Bot. 1905, pp. 76, 202), considered by him too indefinite and unsatis- factory ; while J. Lange's B. macrothyrsus (which we have treated as a synonym of B. gymnostachys) takes its place, and becomes a numbered " species." He says, with reference to B. gymno- stachys, " Vidi specimina Genevierii a B. macrothyrso non dis- tinguenda, sed botanici Gallici auctorem diversos Eubos sub hoc nomine comprehendisse asserunt." And, again, under B. macro- thyrsus, he adds, " Aspectus hujus plantae ab illo B. vestiti valde discrepat, sed notae ditferentes vacillant. E ramis exsiccatis facile ' species ' artificiales construuntur, sed specimina Britan- nica, Hersynica et Holsatica satis congruere videntur." With us the name gymnostachys has certainly not hitherto been applied to one form only, but has included slightly varying forms of B. leuco- stachys, together with our strongly marked and highly glandular Bangor (Carnarvon) plant, which is clearly indistinguishable from Lange's B. macrothTjrsus as supplied to me from " near Kiel, Holstein," by Messrs. Friderichsen and Gelert. Quite the same plant was found in Northants by Mr. Druce last year, and my herbarium contains several other sheets from English counties — from Yorks and Flint to Somerset, Dorset and Devon. So the name gymnostachys may well be dropped from our Bubus list, and macrothyrsus substituted for it, either (preferably, I think) as a strongly marked variety of B. leucostachys, or in an independent position, as in Bubi Europm. B. leucanthemus P. J. Muell. ? There seems to be no reference 204 THE JOURNAL OP BOTANY to this in Rubi Eiiropm except as a synonym of " -B. vestikis," with the note " Frutex B. vestiti unicus in vicinitate oppidi Weissen- burg observatus sub hoc nomine descriptus est." Perhaps some of the plants for which I have from time to time suggested this name would be better placed, as varying forms or hybrids, under li. hypolcuciis ; but most of them from widely separated counties — Westmoreland to Dorset in the west, and Sussex, Kent and Surrey in the east — agree closely with the greater part of Genevier's description of B. hucanthemns Miill. !, as reported by him from several districts in France. So there seems good ground for the provisional retention of this name in our list, either in an independent position between B. hyi)oleucus and B. leucostaclujs, or as a variety (or possible hybrid) under one of them. B. claniciis Focke. Dr. Focke has found himself obliged to alter this name, as he explains " nomen erroneum ; planta a botanicis Danicis quidem lecta sed nondum in Dania ipsa reperta est." He now calls it B. orhifolius (Lefvre, exs.) Boulay in Eouy et Cam. Fl. Fr. vi. 22"; and keeps it, as before, near B. macro- phyllus. He still regards our British plant as one with it : — '• Sowie zerstreut in England and Wales." B. mollissimus Eogers, which in our list appears (after " B. danicus ") as a variety under B. hirtifolius, is placed as a synonym for " B. S2ihcanus P. J. Muell. in Boulay, Konc. Vosg. p. 34, no. 27 (1866)," though the only distribution mentioned, is " Zerstreut im Westlichen England, in Wales und Irland," in exact agreement with that given for Diollissivms in my Handbook, p. 49. B. egregins Focke, var phjmcnsis Focke. nov. var. This is an addition to our Flora, the description added to the above name being as follows : — " Eglandulosus vel glandulas brevissime stipi- tatis in caule, inflorescentitl vel bracteis sparsas gerens. Caules ssepe sat dense pilosi ; inflorescentia variabilis, nunc angusta longa subracemosa, nunc vario modo composita. — Britannia." A further reference to it later is " Die var. phjmcnsis in sildlichen England, besonders aus dem Tale des Plym und an andern Stellen bei Plymouth bekannt." Dr. Focke wrote to me about this plant two years ago as follows : — " Once I gathered, with Mr. Briggs, a late-tlowering, small Buhus in the Valley of the Plym. In general aspect it was near B. longitJiyrsiger, being in fruit at the same time ; but the stronger stems of the unnamed plant were of a more upright growth. Lately I examined my specimens anew, and I could find no essential difference from small forms of my B. cgregms. Usually this latter species bears some scattered glands, whereas the Devon plant is eglandular." Probably I have not seen this "Devon plant" growing, and should have failed to place it among our glandular Egregians if I had come across it. Typical B. egregins, so far undetected in Britain, has always proved a difficult species to me ; and though I have long had a good series of specimens, I have so far found myself unable to understand it. The account of '' phjmcnsis " is followed by the following note : — " (Forma : effeminatus Focke, nov. nom. ; molli- ter pilosus inflorescentite extra axillares longte, multiiJorte ; sepala NOTES ON UR. focke's rubi europ^i 205 longe acuminata, in Acre patentia in fructu reflexa ; stamina stylis multo breviora. Fruticem vidi unicum baud procul ab Oxford Britanniae)." I can throw no light on this note, such as seems necessary before the name can be admitted into our list. Some of the other more difficult plants in our " middle and collective group " Egregii come in for original and interesting treatment. Thus under R. mucronatus Blox. (p. 189) we find the note : " Occurrunt vero, praecipue in Britannia media et meridionali, formai complures ambiguae, qute bine inde species constantes aemulantur, praecipue : — " B. mucronatus var. mulicauUs Rogers Handb. Brit. Rub. (1900) 56, efc formae copiosius aculeolatae : ? B. oigocladus (cit. Muell. et Lefvr.) Rogers I. c. 65 ; non Muell. et Lefvr. ex Sudre. " ? i?. Newbouldii (cit. Babingt.) Rogers I. c. 66. " B. Bloxamianus Golem, ex Rogers I. c. 66. " B. regillus A. Ley Journ. Bot. 1896 p. 217 ; Rogers I. c. 67. " Omnes h» plantae accuratius vivae et in locis natalibus examinandae sunt." But after many years' study of these plants, living and dried, I am confirmed in my conviction that all of them, except my var. niidicaulis, are really best placed where they are found in my Handbook and in London Catalogue, ed. x., among (not Egregii but) Eu-Badulce. As for my nudicaulis — an abundant and very constant plant throughout S. Dorset and S. Hants, and reaching I. Wight and S. Wilts — there seems no room for doubt that its closest relation is to B. mucronatus, though it keeps quite dis- tinct from that type. Next we find, on p. 190, as an example of " Formae et pro- species B. mucronato affines " : — " B. Briggsii Blox. in Journ. Bot. vii. p. 33 (1869) teste Archer Briggs Fl. Plymouth p. 125, qui B. Briggsii varietatem B. fusco- atri (ex sensu Briggsiij esse dicit. B. fusco-ater (cit. Weihe), Briggs Fl. Plym. p. 124. B. oigocladus (cit. P. J. Muell. et Lefvre?) Rogers Handb. Brit. Rub. p. 65." This seems a wholly unsatisfactory arrangement. We clearly cannot accept B. Briggsii as the name for our widely distributed and locally common plant, the " B. oigocladus Muell. & Lefv. '? " of my Handbook and of London Catalogue, ed. x. Dr. Focke's explanation is as follows : — " Bloxamii B. fusco-ater (secund. specim. exsicc.) planta erat B. fusco-atri Weihei similis. Jure igitur auctor B. fusco-atrum Briggsii nomine novo salutavit. Briggsius specimen a Bloxamio nominatum a suo (falso) B. fusco- atro distinguere conatus est et demum sub titulo levis varietatis segregavit. Bloxamius vero non unicum specimen sed integram speciem nominare voluit." But in Fi. Plym. p. 124, Briggs gives " b. Briggsii " as a " very rare " " variety " of his " B. fusco-ater Weihe," and states definitely that whereas " Bloxam maintained that it was a distinct species " he " preferred to consider it a variety of B. fusco-ater, as does Professor Babington." I may add that Briggs left me all his Bubus specimens, and that I still 206 THE JOURNAL OP BOTANY have all his Briggsii sheets. He told me in later years that he found it apparently dying out, and so was disposed to regard it as "possibly a quite abnormal form." To me it has looked like a csesian hybrid, and for that reason I have omitted it from my Handbook and from successive editions of the London Catalogue. His '' fusco-ater " {" B. oigoclad^is ? " oi Handbook and London Catalogue) is certainly a very different looking plant, and is not only locally abundant in England but also extends to Ireland, and in a modified form to the Channel Islands. It was Dr. Focke who first suggested the name " B. oigocladus " for it as " probably right," and to that view he held for several years. If he is right, as he probably is, in now declining to accept our plant as M. & L.'s B. oigocladus, it should, I suppose, either remain in our list for the present as " i?. oigocladus Eogers (non M. & L.)," or be altogether omitted until a more satisfactory con- clusion is reached. There seems to me no room for doubt that I have rightly placed it among Eu-Badulce, where B. Briggsii could hardly be put. Dr. Focke's recent treatment of B. anglosaxonicus Gelert and B. uncinatns P. J. Muell. was referred to at some length in Journ. Bot. 1905, pp. 76, 77, 203, and in their case little need be added here. He still holds that B. anglosaxonicns may come under B. ajncidatus Wh. & N.; but he adds (p. 231) the following notes: — "Species potius collectiva quam limitibus definita videtur, sed quamvis planta valde variabilis sit, specimina typica e terris longe distantibus ssepe inter se optime congruunt. Praeter banc formam satis constantem Eogers /. c. distinguit varietates : curvidens A. Ley, vestitiformis et raduloides. Speciei notas atque limites bene exposuit O. Gelert I. c. sub B. anglosaxonico. Haud raro vero difficile est specimina sicca a B . raduld et acanthode distinguere." B. melanoxylon Muell. & Wirtg. This name, given us in the first instance by Dr. Focke {vide Journ. Bot. 1897, p. 47) for a plant widely distributed in Scotland, and occurring in several Welsh and English counties, is now [B^tbi Eurojjcei, p. 216) withdrawn by him as incorrect, and " B. furvi- color Focke nov. nom." substituted for it, with a short description and the statement that it is " like B. melanoxylon, but differs in its concolorous leaflets." None but British localities are given for it. When we come to the Eu-Badulce — i. e. to the Badula as strictly and rather narrowly limited in my Handbook (p. 5) — we find substantial agreement with our names. But a place is found for my subsp. anglicanus under B. macrostachys P. J. Muell., as a near ally of B. radula — an arrangement for it with which I have no quarrel, though on the whole I still prefer a closer association of it with B. radula itself, together with echinatoides, my other subsp. under the species radida, now left so by Dr. Focke as " var. vel ex Eogers subsp." Both these plants, easily recognized and very widely distributed in England and Ireland, are quite obviously less distinctly eu-radulan than B. radula itself, but they seem most helpfully placed in the closest NOTES ON DR. FOCKE's RUBI EUROP.EI 207 possible relation to it, as the alliance is such that at first they were not distinguished from it. As regards the plants which make up our groups Sub- Koehleriani and Sub-Bellardiani, Dr. Focke's treatment of them hardly calls for comment here except perhaps in two instances — phyllo thyrsus and cavatifolius. Of var. x>hyllothyrsus (Frider.) — under B. Bahingtonii Bell Salt. — he now writes : " B. phyllo- thyrsiis Friderichsen, quern Eogers varietatis titulo sub B. Bahing- tonii describit, ex mea sententia, nil nisi B. chlorothyrsus est. Planta anglica sic nominata parum a B. Bahingtonii differre videtur." It evidently must not be concluded from this that in Dr. Focke's opinion his B. chlorothyrsus is a near ally of our B. Bahingtonii, seeing that he describes them in different groups, with fifty-six pages between them ; but his contention is that I have been mistaken in applying Friderichsen's name to our plant, which he regards as only a variety of B. Bahingtonii. I now agree with him in this. In spite of a very remarkable general resemblance, the two plants are so widely different in glandular and acicular development that they must be assigned to different groups — Friderichsen's remaining where Focke places his B. chlorothyrsus, among the Silvatici and immediately after B. silvaticus, while the right place for our plant is among the Suh-Koehleriani and near B. Bahingtonii. It should perhaps be added that the mistake, only now detected, was due to the fact that Mr. Friderichsen's friend, the late Mr. 0. Gelert, when stay- ing with me in 1897, the year after the publication of B. 'phyllo- thyrsus, gave that name to our plant without hesitation, in spite of the difference in glandular development observed at the time. Since then I have from time to time placed with it a considerable number of plants from Surrey and other counties, none of which can rightly claim a place among Silvatici. The name phyllo- thyrsus must, therefore, be removed from our list. It seems equally clear that B. festivus Muell. & Wirtg. should be restored to it (vide Journ. Bot. 1893, p. 45). It might be placed next after B. Bahingtonii, though not as a variety of it, and include several of the plants hitherto named phyllothyrsus. In Buhi Europai it is given as " species conjungens collectiva," and under " var. cu- festivios " we find Mr. Ley's Dinmore Woods (Heref.) plant (No. 95, Set of British Buhi) quoted for it. Perhaps a pro- visional place should also be found, among the Silvatici, for B. chlorothyrsus Focke, in agreement with the note under that name on p. 177 of Buhi EuropcBi : " In Britannia vidi formam foliolis subtus cano-virentibus distinctam (legit W. E. Linton in Derbyshire) quam praeterea exsiccatam a B. chlorothyrso separare non possum." Our B. cavatifolius P. J. Muell., which Dr. Focke apparently accepts as Mueller's plant, he now places under B. pallidas Wh. & N. as " forma luxurians densiflora et latifolia"; but it seems decidedly nearer to our B. Bahingtonii, and (as I have seen it in fair quantity in Herefordshire and Monmouthshire and from two localities in W. Gloucester) clearly and constantly distinct from both. 208 THE JOURNAL OF BOTANY On reaching the last three groups in our list, including all, or nearly all, the Eubi with very mixed armature, the difficulty of comparison with the concluding pages of Buhi Europai is seriously increased. This special difficulty arises partly from the greater range of variation, which appears to be a genei'al feature of the more glandular Eubi, and partly from the circumstance that these highly glandular Eubi with very mixed armature are, as a rule, much more thinly distributed than the members of the earlier groups — in Britain at all events, and probably in most parts of the Continent also. On these plants, therefore, I propose adding only a few brief notes of a more general character. Thus, if we consider them in the order in which they stand in our London Catalogue, ed. x., it may be said in general terms : — 1. The following species are now recognised as common to Great Britain and the continent of Europe: — B. rosaceus Wh. & N., B. hystrix Wh. & N., B. Koehleri Wh. & N., B. viriclis Kalt. (sensu ampliss.), B. Bellarclii Wh. & N., B. hirtns Waldst. & Kit. (sp. collect.), B. dumetorum Wh. & N. (sp. collect.), B. diversi- folius Lindl. (a species in Bubi Eiiropcei, but a var. in London Catalogue, ed. x.), B. corylifolius Sm., B. Balfourianus Blox., B. ccEsius Linn. These all are strongly marked plants, and may be regarded as somewhat widely distributed here and on the Continent. 2. Less strongly marked ; as a rule less widely distributed, and therefore somewhat more open to doubt as to their identity here and on the Continent, and perhaps partly on that account not reported for Britain in Biibi Europcei, though probably cor- rectly included in our list, are the following: — B. adornatus P. J. Muell., B. horridica2ilis P. J. MuelL, B. hostiUs Muell. & Wirtg., B. fusco-ater Weihe, B. divexiranms P. J. Muell., B. serpens Weihe, B. Kaltenhachii Metsch., B. mimUiJiorus Wirtg. (or P. J. Muell. ?), B. saxicolus P. J. Muell., B. tereticaulis P. J. Muell., B. cyclophyllus Lindeb. All these names (except perhaps B. cyclo- 2)hyllus, which was first suggested as British by Babington) have from time to time, and mostly more than once, been given to British plants by Dr. Focke, though there seems no evidence in Bubi Europai that he would still so apply them. As regards cyclophyllus, we have followed Babington in substituting this name for his conjungens, to be used in an aggregate sense for forms of^ B. corylifolius Sm. other than the type (" a. sublustris Lees " of London Catalogue, ed. x.). As however, there seems to be considerable uncertainty about the right use of Lindeberg's name, it may be wise to substitute conjungens Bab. as is done in Bubi Europm. 3. British plants for which names different from those in our list are preferred in Bubi Europcei : — B. plmthostylus Genev., B. Marshalli Focke & Eogers and its " b. semiglaber Eogers," B. acutifrons Ley. Brief notes on these may be found of interest. B. plintliostylus Genev. is given in Bubi Europai as a synonym of B. Beuteri E. Merc, but our plant, apparently known only AZOLLA IN BRITAIN AND IN EUROPE 209 from our descriptions, is not accepted as certainly the same. Our R. Marshalli Pocke & Rogers is placed as " var. Marshallii Focke & Rogers " under li. pilocarpus Gremli, together with our very variable var. semiglaher. On this Mr. Marshall's translation (Journ. Bot. 1905, p. 77) of Dr. Focke's remarks in Aschers. & Graebn. (1902-3) may be consulted. Apparently our very re- markable type has not yet been found on the Continent. As to the treatment of B. aciUifrons Ley, reference may be allowed to p. 204 of the same volume of this Journal, where reasons are given for hesitation in accepting the proposed substitution of B. humifasus Wh. & N. for Ley's name. Ley's var. amplifrons (Journ. Bot. 1902, p. 69), though a very strongly marked form, does not seem to be noticed in Buhi Europcei. 4. Described in Buhi Europai as British, but not as yet detected elsewhere in Europe : — B. Purchasianus Rogers, B. cocj- natus N. E. Brown, B. dasyphyllus Rogers, B. ochrodermis Ley. B. cognatus has a distinct place and brief description in Bubi Europcei, but is held to be very like the Scandinavian B. horridus Hartm. B. dasyphyllus is very briefly and inadequately described by Dr. Focke, and as British only ; but in 1912 I received excel- lent specimens of it collected by Mr. Friderichsen " in woods, Tolne, Denmark," in that year. Apparently otherwise still un- known on the Continent, though exceedingly abundant in Great Britain and frequent in Ireland. 5. British plants included in London Catalogue, ed. x., but apparently not mentioned in Buhi Europcei : — B. Poioellii Rogers, B. Durotrigum R. P. Murr., B. rotundifolius Bab., B. Buchnalli J. W. White. 6. British plants recently published, but not included either in London Catalogue, ed. x., or in Bubi Europai : — B. lacustris Rogers, in Journ. Bot. 1907, pp. 9, 10, B. glareosus Rogers & Marshall, in Journ. Bot. 1912, pp. 309, 374. AZOLLA IN BRITAIN AND IN EUROPE. By a. S. Marsh, B.A. [Reprinted by permission from the Proceedings of the Cam- bridge Philosophical Society, vol. xvii. part 5 (May 5, 1914).] In the middle of October 1913 a species of AzoUa was found in Jesus Ditch, Cambridge, by Mr. H. Jeffreys, of St. John's College. Mr. Moss called my attention to the fact, and at his suggestion and with his frequent kind assistance I have identified the species and collected a few notes on the distribution of plants of this genus in Europe generally and the British Isles in particular. The Cambridge plant I found to be Azolla filiculoides Lam. It was growing among the Lemna, but two or three large patches, several metres broad, bore Azolla almost pure, the dull brownish colour of the plant as seen in large masses showing up markedly Journal of Botany. — Vol. 52. [August, 1914.] r 210 THE JOURNAL OF BOTANY against the bright green of the duckweed. When first found the plants seemed to be without reproductive organs, but on November 2nd it was bearing micro- and macro-sporocarps in some quantity. On November 26th, after several sharp frosts, the Azolla was growing vigorously, still with sporocarps, and had spread over larger areas, at the eastern end of the ditch becoming the dominant species of the aquatic vegetation. At the present time (February 9th) it is very abundant, but very red in colour and broken up into small pieces. As to means of introduction of this fern into Cambridge we are completely ignorant. The nearest of the previously recorded stations is the Norfolk Broads area, while the obvious suggestion, that we are dealing with a Botanic Garden escape, is untenable, since there was before this discovery no Azolla except A. caro- liniana being grown at the Cambridge Botanic Garden. Azolla, according to Baker,'' is a genus with five species inhabiting the tropics and warm temperate regions of both hemi- spheres. Of these species two have been introduced into Europe, and both occur in the British Isles. These two are A. caroliniana, which occurs native in America from Lake Ontario to Brazil, and A. filiculoides, from South America. I The characters of these two species have been well summed up in two recent papers on the occurrence of A. filiculoides in Europe, and from the accounts of these authors (viz. Bernard | and Beguinot and Traverso§), from Baker ij and from von Martins, *! the following details of the principal differences between the species are taken. Azolla filiciiloides (Lamarck, Encyclopedie Methodique : Botanique, T. i. p. 343 and plate 863, 1783). The plants are in dense tufted masses, the ends of the shoots being porrect and often protruding, not lying fiat on the surface of the water as in the other species. The whole shoot is much larger and thicker, the branching is more compound and the branches are closer together. The upper lobes of the leaves have a broad distinct margin, and bear numerous unicellular trichomes on their upper surfaces. The reproductive organs show the most distinctive characters. The glochidia or hooked hairs which are attached to the massulcz or microspore masses have non-septate stalks. The macrospore wall is furnished with large, deep, circular pits. Azolla caroliniana (Willdenow, Species Plantarum, v. p. 541, 1810). The plants are much smaller with much less dense branch- * Baker, Fern Allies, p. 137, London, 1887. t The distributions are as given in Coste, Flore de France, iii. pp. 702, 703, Paris, 1906, and Ascherson u. Graebner, Synopsis der initteleuropilischen Flora, i. p. 114, Leipzig, 1896. \ Bernard, Recueil des Trav. Bot. Neerland., i. pp. 1-14, 1904, quoted in the Report of the Botanical Exchange Club for 1912, p. 186. § Beguinot e Traverse, " Azolla filiciiloides Lam. nuovo inquilino della flora italiana," Bull. Sac. Bot. Ital., pp. 143-151, 1906. || Baker, loc. cit. *'^ von Martius, Flora Branliensis, vol. i. part ii. p. 657, plate 82, Leipzig, 1884. AZOLLA IN BRITAIN AND IN EUROPE 211 ing. They lie flat on the surface of the water. The roots are not as numerous or as conspicuous as in A. filiculoicles. The margin of the upper leaf lobe is not as broad as in the other species, and the trichomes of the upper surface are said to be bicellular, though I have not been able to observe this character satisfactorily. The glochidia have 3-5 transverse septa in the stalk, and the macrospore wall is not pitted but merely finely granulate. The history of the genus in Europe began in 1872, when A. caroliniana was introduced into Continental botanic gardens, whence it soon escaped into neighbouring ditches and ponds, and multiplied enormously. In 1878 De Bary described it as a " new water- pest " in Kassel, and in 1885 it was very abundant at Leyden and Boskoop in Holland. * It was also found at Bonn, Giessen t and Strassburg ^ in 1885, and in Berlin in 1887.]: In Bohemia it was found by Celakovsky near Pilsen in 1895, and it had spread much earlier into England (1883), France (1879), and Italy (1886). § In England A. caroliniana was first obtained at Pindon (Middlesex), and an account of this is published in Science Gossip for 1883. It has been recently reported from various spots in the Thames valley, between Oxford and London, but it must be remembered that until Ostenfeld ji pointed out the fact in 1912 (from specimens found in 1911) it was not realised that we had any Azolla other than A. caroliniana. For instance, DrucelT (1908) gives only one species, A. caroliniana. The following records for the British Isles have been published, though, until the material has been re-examined in the light of Ostenfeld's discovery, they must be considered recoi'ds for the genus rather than for the species. Azolla described as A. caroliniana has been found at Hayes Place (Kent), Oxford, Sonning, Henley, Enfield, Sunbury and Suleham.*" Of these I have been able to examine material from Sunbury and Enfield kindly sent by Mr. 0. E. Britton. The Sunbury plant is ^. filiculoicles, the Enfield specimen A. caroliniana. Another Azolla from Enfield was sent by Mr. HoUoway, but this was A. filiculoides. The Norfolk Azolla, which is good A. filiculoicles, has also been several times referred to as A. caroliniana. I have seen A. caroliniana from one other British locality, viz. Godalming, where it was found in 1913. The species is described by Ascherson and Graebner (1896) * Kittel, Gartenfiora, 1885. t Dosch u. Scriiia, Excursionsjlora Hessen, 3te Auflage, p. 24. I Luersseii, Farnpflanzen, p. 598. § This account is taken chiefly from Ascherson and Graebner, loc. cit., but see also Saccardo, Cronologia della Flora I(aZ/(i;!rt, Padova, 1909 ; Ibid., " De diffusione Azollie earoliuianffi per Europam," Hedwigia, 1892, p. 217; Beguinot and Traverso, loc. cit., where many additional references are given. Ij Ostenfeld, " Floristic Eesults of the International Excursion," New Phyt. xi. p. 127, 1912. H Druce, List of British Plants, p. 88, Oxford, 1908. ♦* Reports of the Botanical Exchange Club, 1910, p. 609; 1911, p. 56 ; 1912, pp. 186, 220; Journal of Botany, xl. p. 113, 1902; xlviii. p. 332, 1910. R 2 212 THE JOURNAL OF BOTANY as fruiting only very rarely, they knowing of only one ease of fruit being produced in Europe — a record from Bordeaux. No fruiting material has been found in the British Isles, although fruiting A. filiculoides has more than once been described under the wrong specific name. Azolla filiculoides was introduced into Europe in 1880 by Eoze, '' who naively remarks, " Le climat de Bordeaux paralt, du reste, assez bien convenir a ces deux esp^ces americaines, car quelques poign6es de la premiere [.4. caroliniana] en 1879, et de la seconde [A. filiculoides] en 1880, jetees qk et la dans les fosses des marais de cette ville, ont donne naissance a une legion innombrable de ces plantes, qui ont envahi presque tous les fosses, mares et 6tangs du departement de la Gironde." It spread over many parts of France and then into other countries. In 1896 Ascherson and Graebner knew of it only in western and northern France. In 1900 it had reached Italy, f In the British Isles A . filiculoides was first noticed as a distinct species by Ostenfeld ]: in 1911, who found it at Woodbastwick, Norfolk, and at Queenstown Junction, Co. Cork. It was, however, present in this country before that time. The Sunbury record of A. caroliniana in 1910 § should certainly be ascribed to the other species, while the Azolla was noticed in the Norfolk Broads || before Ostenfeld's identification. I have also seen fruiting specimens of A. filictdoides found in 1912 at Almondsbury, West Gloucestershire, and kindly sent me by Miss I. M. Roper. The same species now occurs at Reading, where it is peculiar in being without the endophytic blue-green alga, AnahcBna, which usually inhabits the cavity of the upper leaf-lobe. At present ^4. /2ZiC2Jo/f?cs seems to be growing in importance as a constituent of British vegetation, for, as the result of the floods of 1912, it has been distributed over large areas in Norfolk. It is described as occupying a definite position as a member of the association of Typha ang^cstifolia, especially in South Walsham and Ranworth Broads. T A. filiculoides fruits quite readily in Europe. Both the specimens found by Ostenfeld were fruiting, the Almondsbury and the Sunbury plants were in fruit, and I obtained fruit last autumn, not only from Cambridge, but, by the kindness of Mr. W. E. Palmer, of St. John's College (the author of the article in Nature), also from Norfolk. Ascherson and Graebner*''' also describe it as a freely fruiting species. * Eoze, " Contribution a I'etude de la fecondation chez les Azolla," Bull, de la Soc. hot. de France, xxx. p. 198, 1883. f Saccardo, Cronologia, loc. cit. ; Beguinot e Traverse, loc. cit. + Ostenfeld, loc. cit. ; Report of the Bot. Exch. Club for 1912, pp. 220, 301. § Re}]. Bot. Exch. Clnb for 1910, p. (i09 ; Joimi. Bot. xlviii. p. 332, 1910. II Rep. Bot. Exch. Club'for 1910 and 1911, loc. cit. M Palmer, "Azolla in Norfolk," Nature, xcii. p. 233, 1913. The plant is wrongly named A. caroliniana, but I have seen fruiting specimens, which prove it to be A. filiculoides. ** Loc. cit., p. 115. NOTES ON THE MANX Fr,ORA 213 In conclusion, I should like to suggest that it is of some importance to keep a look-out for Azollas in the British Isles, as in the event of their becoming important factors in our vegetation, as full a knowledge as possible of their early history in the country would be of great interest and value. NOTES ON THE MANX FLORA. By J. W. Hartley and J. A. Wheldon, E.L.S. Whilst engaged in making a lichenological survey of the Isle of Man, early in June of the present year, we noticed a few flowering plants, some of which may prove to be new to vice- county 71, or at least from new stations. But there is great need of a revision of Manx plants, and it is hoped that a general botanical survey of the Island may be undertaken shortly. The most recent lists of species that we are aware of are the Rev. S. A. P. Kermode's " List of Flowering Plants," reprinted from Yn Lioar Manninagli, vol. iii., in 1900; and the " Flora of the Manx Curraghs," by one of us, which appeared in the Lancashire Naturalist, 1910, pp. 271-274 and 301-304. Mr. Kermode's Hst does not include the GraminecB or Gijperacece,, and specific loca- lities are only quoted for the rarer species. Lists of Grasses and Ferns have, however, appeared in local publications, and in 1889 the Rev. S. Gasking contributed to Research a catalogue of Manx plants without localities, which included the Grasses, Sedges, and higher Cryptogams. Gheiranthus Cheiri L. On walls, Rushen Abbey. Cochlearia danica L. Shore near Glen Maye. Reseda Luteola L. Copse near Jurby. Viola ericetorum Schrad. Heathy ground near Blue Point. Polygala oxyptera Reichb. Sand dunes about Jurby and Blue Point. Gerastium vulgatum L. North of Peel. G. viscosum L. Shore near Blue Point. G. tetrandruvi Curt. Scattered, with G. semidecandrum L., on the dunes from Jurby to Blue Point. Arenaria serpyllifolia L. var. viacrocarpa Lloyd. With the preceding species. Spergularia rupestris Lebel. Peel ; and very fine on rocks on the Glen Maye shore. Sagina ciliata Fr. Sparingly near Point of Ayre. Erodium cictUarium L'Herit. var. glutinosum Clav., and E. maritimum L'Herit. On the dunes near Lhane. Ononis maritima Dum. (0. repens L. wav. prostrataBveh.). A very glutinous prostrate form at Jurby and near Point of Ayre. Lotus corniculatus L. var. crassifolius Pers. Shore near Blue Point. Rosa spinosissima L. Dunes near Lhane. A taller state is frequent on hedge-banks about Jurby and Blue Point. 214 THE JOURNAL OP BOTANY Sechim anglicum Huds. Abundant on the dunes of the north_ where it seems to replace S. acre, which was not seen. Smyrnium Olusatrum L. Ballasalla, Jurby, Lhane, &c., very frequent, and nearly always affected by Puccinia. Anthriscus vulgaris Bernh. Lhane and Jurby. Galium verum L. var. maritimum DC. Sand-dunes near Lhane. Valerianella olitoria Poll. Common on the sand-dunes from Jurby towards Point of Ayre. Dij^sacus sylvestris Huds. Seen between Glen Cam and Jurby, but the locality was not noted. We find, however, that it is not included in Mr. Kermode's list. Petasites fragrans Presl. Bishop's Court Glen ; no doubt planted. Cnicus arvensis Hoffm. var. vestitus Koch. Near St. Germans. Taraxacum Icevigatum DC. Sand-dunes near Lhane. Jasionc montana L. A small form with prostrate stems, and anthodes as small as those of var. littoralis Fr., but with hairy leaves, occurs on the dunes near Jurby and Lhane. Euphrasia curta Wettst. A very dwarf form on the dunes at Lhane. Salix aurita L. and S. lutescens Kern. Glen Kushen. Polygonum Rail Bab. Near Blue Point. (Eecorded as P. Bohcrti Lois, in Lanes. Naturalist, 1910, p. 304.) Scilla verna Huds. Very fine on the coast about Glen Maye. Carex arenaria L. Common on the dunes about Jurby and Lhane. C. flava L. Boggy fields near St. Germans. Phleum arcnarium L. Plentiful at Blue Point and Lhane. Ammopliila arenaria Link. Sand-dunes from Jurby onwards to Point of x\yre. Aira car yopJty Ilea L. and A. prcecox L. Both abundant in the north of the Island, on sand-dunes and hedge-cops. Trisetum flavcscens Beauv. Glen Maye. Festuca rigida Kunth. Peel Castle. F. rottboellioides Kunth. Rocks on the shore near Glen Maye. Bromus hordeaceus L. var. leptostachys Beck. Dunes near Jurby. BotrycMum Lunaria Sw. Amongst Pteris, on the dunes near Blue Point. ARMERIA ARCTICA Wallr. FOSSIL IN BRITAIN. By Clement Reid, F.R.S. Among some fruits associated with Salix polaris in a deposit belonging to the Glacial Period, discovered at Borna, south of Leipzig, Dr. C. A. Weber has recently found and figured Armeria arctica Wallr., a species supposed to be confined to Arctic North America and Greenland. He, however, found the same plant in A NEW SPECIES OF DISCINELLA 215 the herbarium at Stockhohn, among specimens from Siberia, where it had not previously been recorded. He also again recog- nized it, among plants wrongly determined, from the glacial deposits of Denmark and perhaps Galicia. ■' Armaria arctica is therefore circumpolar, and it had formerly a wide extension southward. Dr. Weber's figures and description made me re-examine some unknown fruits which I had obtained at various times from British glacial deposits, but could not determine ; they certainly did not belong to A. maritivia, and the American A. arctica was not in my collection. Comparison leaves no doubt that this circumpolar species was common also over the lowlands of Britain during the Glacial Epoch. I have fruits of it from the base of the boulder clay at Mundesley, on the Norfolk coast (associated with Salix polaris and Betula nana) ; from the top of the glacial deposits at Saughton and Corstorphine, close to Edinburgh (associated with S. polaris, S. herbacea, S. reticulata) ; and from material collected by Messrs. E. T. Newton and S. H. Warren at Bonder's End, in the Lea Valley, where also it is associated with Salix herbacea and Betula nana. As this Arctic species was formerly so widely distributed over the lowlands of Europe, the living mountain forms of Armcria in Britain should now be re-examined critically, for it is probable that they will be found to belong to this species, or to one of the allied Arctic species, and not to the sea-coast A. maritima. As distinguished from A. maritima the fruiting calyx of A. arctica is more robust, shorter, broader, much more openly campanulate, and is densely pilose on the ribs but smooth on the intervals. All these characters are well seen in the fossils, though some of the specimens, as we should expect, have lost the hairs. The smooth intervals at once separate A. arctica as belonging to a different section of the genus from that in which A. maritima is placed. t The fruiting calyx gives the best specific characters in this genus. A NEW SPECIES OF DISCINELLA. By J. Ramsbottom. The genus Discinella was founded by Boudier in his " Nouvelle classification naturelle des Discomycetes charnus," in Bull. Soc. Myc. Fr. 1885, i. p. 112. There is no Latin description, but the generic characters can be easily understood from the information supplied in the key to the genera, and from the fact that Phialea Boudieri Quel, is made the type species. The characters are * Die Mammutflora von Borna, Abh. Nat. Ver. Bremen, 1914, Bd. xxiii. Heft 1. t See Boissier in De Candolle, Prodromus Systematic Naturalii' Herini Vege- tabilis, pars xii., 1848. 216 THE JOURNAL OP BOTANY more fully given by the same author in his Histoirc et Classifica- tion des Discomycetes cVEurope (1907). According to Boucher the principal characters are the terrestrial habit and the size of the fungus, which can reach as much as 12 mm. The species are thick, shortly stipitate, having somewhat the appearance of oper- culate species, though they are inoperculate. The hymenium is more or less surrounded by a dentate margin. The exterior of the receptacle is subtomentose. The inoperculate asci are remarkably small, eight-spored, with a raarginate pore. The paraphyses, which are fairly slender, are filled with oil globules, united some- times into masses. The spores are fusiform and contain oil drops, which are accompanied or not by granulations. (The genus Discinella of Karsten {Hedwigia, 1891, 30, p. 301) is different from Boudier's genus : " est Discina Fr. em. apotheciis minoribus," and the species described, D. corticaUs, is not terrestrial.) The species of Discinella which have been recorded in this country are D. purpurascens (Pers.) Boud., D. exidiiformis (B. et Br.) Boud., and D. Menzicsii Boud. The last-named species was described and figured in Trans. Brit. Mycol. Soc. 1913, iv. p. 62, as Calijcella Menziesii, from specimens sent from Perth by Mr. J. Menzies. Boudier, in the same publication, 1914, p. 323, places the species in the correct genus. At the end of last year Mr. D. Garnett found this fungus at Silchester, growing amongst moss and usually under Ulex, and brought it to me for identification. As the fungus appeared suitable for cytological investigation, several collections were obtained, and an attempt was made to find young stages. In the search numerous very small bodies were met with which, on examination, proved to be mature apothecia. These specimens were kindly brought to me for examination, Mr. Garnett fully realizing the close similarity these fructifications had to those of 1). Menziesii. The fungus is apparently new, and is distinguished from its allies by the exceptionally small size of the apothecia. It would be included by Saccardo in the genus Humaria. Discinella minutissima Ramsb. et Garn. — Minutissima sub- stipitata, isabellina, 300-400 /x lata, carnosa, crassa, margine non prominulo, obtuso ; stipite obconico, crasso ; paraphysibus tenui- Idus, 2-3 /x crassis, oleosis, simplicibus, filiformibus, ad apices non aut vix incrassatis, hyalinis ; ascis inoperculatis minoribus octo- sporis, foramine marginato, cylindrico-clavatis, 40-50 /x x 4-5 jj., ad basim vix attenuatis, ad apicem iodo non caerulescentibus ; sporidiis oblongo-fusiformibus, hyalinis, levibus, continuis, stepe leniter cur- vatis, 7-8 /x X 2fx, intus guttulosis et granulosis, guttulis SEepius 3 majoribus, granulis minoribus. Ad terram argillosam in Silchester, Hants, x\pril 22, 1914. 217 SHORT NOTES. EoMULiEA PARViFLOBA (p. 46). — Dr. Stapf calls my attention to the publication of this name in Bubani's Flora Pyrcnaa, iv. 150 (1901). Judging from the synonymy, this is identical with Salisbury's Ixia iMrviflora, and the name must therefore be cited as of Bubani. It is curious to note that Bubani does not cite Salisbury's publication, of which he was probably not aware ; he based his name on " /. parviflora Pourr. Herbr. Matr.," to which he adds " ita ego." — James Britten. Apera interrupta Beauv. in Lancashire. — Mr. J. A. Wheldon and I found this rare grass at Freshfield, South Lan- cashire, v.-c. 59, early in July. The plant occurred on sandy and cindery soil for some distance along the margin of the road leading from the station to the shore. The specimens were numerous and well-grown. No other "alien" plants were seen in the vicinity, but there can be no doubt that the plant in this locality is purely a casual, as to the origin of which we can make no suggestion. There is no record to our knowledge of this species ever having been met with in Lancashire before, although its congener Apera Spica-venti occasionally crops up about docks, canals, and railways ; and as there is a possibility that it may establish itself on the open sandy soil of the district, it may be well to put particulars of our " find " on record. — W. G. Travis. Miller's ' Abridgement,' ed. 4. — As there seems to be some obscurity about the " discovery " of the fourth edition of the Abridgement (1754) of Miller's Gardeners Dictionary, I may point out that I was the first to call attention to it as an autho- rity for the citation of genera, in Journ. Bot. 1910, p. 183, and again in Prodr. Fl. Brit. p. 461 (February, 1911). The many binomials to be found in it are, of course, only accidental and are not valid for citation ; and have even less claim to notice than similar " accidentals " in Hill's British Herbal and in Gersault's List, which are both of later date. As it has been suggested that this work of Miller may be ignored by decision of Congressional vote on the principle of the "nomina conservanda" anomaly, I should protest that it can be no more suppressed than the classical but badly printed Species Flantarmn of the previous year. — F. N. Williams. REVIEWS. Physiological Plant Anatomy. By Dr. G. Haberlandt. Translated from the fourth German edition by Montagu Drummond. 777 pages, with 291 figures in the text. Macmillan & Co., Ltd. 1914. 25s. net. The relation of an organism to its environment is one of the most interesting of all biological problems if, indeed, it cannot be said to comprise the whole of biology. The "hand and eye " 218 THE JOURNAL OF BOTANY characters of plants, the various macroscopic adaptational devices which they display, must at times appeal to all botanists. For those who are fortunate enough to have had a training in micro- scopic manipulation there is something further — the possibility of the attempt to correlate the internal anatomy of plants with their life-processes and with their surroundings. The study of plant anatomy as such is of purely academic interest. It is only to a certain type of mind or, in any case, only to a trained anatomist, that the various forms of " woody fibre " and their position in the plant can appeal. The older observers in anatomy and physiology certainly realized the functions of many types of plant tissue. Anatomy and physiology w'ere, however, kept distinct. An epoch- making paper by Schwendener — " Das mechanische Prinzip im anatomischen Bau der Monocotylen " — was published in 1874, in which the skeleton of the plant and the structure and relation of this " mechanical" system were correlated with function in a very convincing manner. Every examination student is now taught the essence of Schwendener's principles but, probably because no controversy was aroused, Schwendener's name is not so often attached to them as it is to that suggestion of his which revolutionized the study of lichens. Many of Schwendener's pupils adopted the anatomico-physiological attitude, and in 1884, Haberlandt (who has succeeded Schwendener as Professor of Botany in Berlin) published the first edition of his famous PhysiologiscJie Pflanzenanatomie. The book under review is a translation of the fourth edition of this work. " In its present form, therefore, this work may be assumed to embody the mature and considered views of its author, with regard to that section of botanical science which he has made peculiarly his own." It is therefore more satisfactory to have a careful translation of this work than a book which is merely a compilation. The volume opens with an introduction of thirteen pages, in which the author defines the aim of his book and attempts to refute certain philosophical objections to parts of the study: — "The object of Physiological Plant Anatomy is twofold. It consists, first, in the recognition of the physiological functions pertaining to the tissues of the plant and to the structural units, or cells, of which these tissues are composed ; and, secondly, in the discovery of the connection that exists between the several functions and the anatomical arrangements required for their proper performance." There is also a discussion concerning " functionless " cells which " play no useful part in the general economy of the plant." This seems rather too definite ; on the other hand, we should be inclined to doubt whether some of the functions bestowed upon certain cells have any real existence. If a function cannot be assigned to a tissue at present, it seems as illogical to assume that no function exists as it does to hold that the functions assigned are in all other cases the correct ones. " The value of teleological explanation depends entirely upon the philosophical attitude of its author " we are told, and in some cases the attitude of the present author is apparent. ETUDES SUR LA FLORE DU KATANGA 219 There are fourteen chapters : — The cells and tissues of plants, meristematic tissues, dermal, mechanical, absorbing, photosyn- thetic, vascular, storage, aerating, secretory and excretory, motor, sensory and stimulus-transmitting systems, and a concluding chapter on secondary growth. Each of these chapters is sub- divided into convenient portions. The translator's part of the work seems to be well and care- fully done. Owing to the adoption of a somewhat free trans- lation, the book hardly reads as if it were translated. The time that has elapsed since the fourth German edition (1909) appears rather long, and meanwhile certain portions of the study, for instance, those on light perception, have been somewhat extended. The recent results have not been incorporated by Mr. Drummond, we think advisedly. The only additions made are indicated by square brackets and seem to consist merely of words added for greater clearness. An innovation is the gathering together of the notes at the end of the book instead of the perhaps more conve- nient place at the end of each chapter which they occupied in the German editions. There is a subject index and an index of plant names. Botanical students will welcome this book. There are few laboratories which do not possess the original work, but the German language is not, as a rule, read by students with ease. The binding and printing are both excellent and the figures clear. The author is to be congratulated on the completion of his arduous task of making this standard work accessible to many who would not otherwise have been able to read it. J. Eamsbottom. Etudes sur la Flore clu Katanga. (Annales die Musce du^ Congo Beige, Botanique, ser. iv. vol. ii. fasc. i.) Par Em. de WiLDEMAN. Brussels, Sept. 1913. Dr. de Wildebian continues his researches into the flora of the southernmost province of the Belgian Congo in a Memoir in which a large proportion of the families of Pteridophyta, Monocotyledones and Dicotyledones occurring in that part of Africa find a place. In a work so discursive it is no matter for wonder if some of the author's determinations invite criticism. For instance, the plant named Geophila herbacea K. Schum. {Kassner, 2427) seems to us to be G. reniformis Cham. & Schlecht. ; so, too, Pentanisia variabilis Harv. is rather P. ScJuoeinfurthii Hiern, and Nicxia platyphylla Gilg is N. sambesina Gilg, while for Buellia prcetermissa Lindau, B. patula Jacq. should be read, and Kassner's 2908 is certainly not Brillantaisia patula T. And., nor is the same collector's 2619 Barleria salicifolia S. Moore ; moreover, Kassner's 2405 is PhyllantJms leucanthus Pax, not as here stated P. odontadenius Miill. Arg. Among several over- sights we may mention that the plant called Vcrnonia Kassneri De Wild. & Muschler must be renamed, the trivial being already occupied, and the same remark applies to Ipomixa Kassneri. The letterpress also is not entirely satisfactory ; thus the headings 220 THE JOURNAL OP BOTANY Myrtacece and Melastomacece have become mutually displaced, and what Phili2)2)ia is doing in the Umbelliferous galley is not evident. Strong objection must be taken to Dr. de Wildeman's practice of publishing specific names without a reference ; thus we find, inter multa alia, reference to Sojjubea Kassneri Pilger " in hb. Berol." for a plant already described, and other determinations similarly authenticated of species still lacking a published description ; the names, too, of several of Mr. Edmund Baker's Crotalarias not yet formally published appear here simply with " Baker sp. nov." after them ; also we are given no means of distinguishing the younger botanist from his venerable father, who must have been a marvel of precocity if he was already writing, as we are told he was, in Hooker's London Journal oj Botany seventy-one years ago ! One is sorry to make remarks of this kind concerning a work not without its good points. The many descriptions of new species are written, as is usual with Dr. de Wildeman, in an admirably clear manner, and the photographic plates show up the species more effectively than is often the case with that style of illustration. But there are too many signs of haste throughout the memoir to enable one, after careful examination, honestly to praise it without many reservations. g_ ^ Two Additions to British Local Floras. A Supplement to the Flora of Somerset. By Edward Shearburn Marshall, M. A., F.L.S. 8vo, cloth, pp. iv. 242. Taunton: Published by the Somersetshire Archteological and Natural History Society. 1914. Price 7s. 6d. Flora Orcadensis : containing the Flowering Plants arranged according to the Natural Order by Magnus Spence, and the Mosses by Lieut. James Grant. 8vo, cloth, pp. xcv. 148. With Maps and Portraits. Kirkwall : D. Spence. 1914. Price 4s. The extent of Mr. Marshall's Siipplement to R. P. Murray's Flora of Somerset (1893-6) confirms the view always held by the present writer that the estimate of Murray's book published in this Journal for 1897 (p. 150) was somewhat too high — it was a good book, but hardly of " first-rate botanical excellence," falling short as it did of the standard raised by Trimen and Dyer's Flora of Middlesex (1869), and mixintained by many other works, of which Mr. White's Flora of Bristol (1912) is the most recent. The Somersetshire portion of the last-named work has been made full use of in Mr. Marshall's book; indeed, he calls it his " main- stay," an estimate which seems a little over-generous, for the Supplement owes its chief value to Mr. Marshall's own notes, and to the greatly extended information regarding the plants of the county and their distribution. In some respects the Supplement is susceptible of improve- ment. There should have been an introduction in which the SUPPLEMENT TO THE FLORA OF SOMERSET 221 principal additions or exclusions, whether of species or of dis- tribution, should liave been specified ; and it would iiave been well if the page of the original work had been indicated under each species included therein and appearing in the present work — the number of the species under each genus might, at any rate, have been cited. One would have been glad to have had some biographical matter, which is entirely lacking in Murray's Flora ; and we think some distinction should have been made between notes— e.(/. those on Berberis aristata and the hybrid Daphne — which originally appeared in this Journal, and those which are now first published. Other references might conveniently have been added — e.g. to Plantago Coronopus var. sabrina — and one is a little surprised to find no acknowledgement in the brief preface of the evidently considerable help given by Miss Eivett and Miss Roper. Turning over the pages, we note much matter of general interest. There are many new localities for Aconitum, which Mr. Marshall thinks (and those who have seen the plant in situ will agree with him) a true native in Somerset : Dr. Stapf, we are told, has " been unable to meet with exactly our English plant on the Continent." The treatment of Viola in accordance with Mrs. Gregory's monograph has led to numerous corrections and additions, extending over seven pages. The " usual if not the only Somerset plant" of the Alchemilla vulgaris aggregate is A. minor Hudson {A. filicaulis Buser). Daucus gummifer, which appeared in the Flora, is withdrawn, as also is Gentiana cam- pestris. The treatment of Hieraciitm is greatly amplified, and contains interesting notes ; the single Euphrasia of the Flora is distributed among fourteen names, the result of observations pub- lished since Mr. Murray's book was written. Mentha also has undergone considerable amplification. It may be noted here that Euphorbia LatJtyrus, which is given in the Flora for Steep Holm as " not truly wild " and " naturalized," was found there by Banks in 1773 ; his specimen in the National Herbarium is endorsed by himself : — " I found this one plant among the Ligustrum on the south side of the Steep Holmes Island, but being hurried by the tide had not time to search for more." There are interesting additions to Salicornia, represented in the Flora only by S. herbacea ; and a long note on the Snowdrop calls attention to the probability of its nativity in the county. The " var. bracteata Druce " of Scilla nonscripta is supposed to be " rather a form than a variety " ; a note to this effect will be found in Journ. Bot. 1908, 200, where it is shown that the authority for the name should be " Hort. ex Baker." Wolffia is an interesting addition to the county list, and Asple- nium germanicuvi was probably accidentally omitted from the Flora. These are but indications of the valuable information which render this Supplement, apart from the Flora to which it relates, of interest to all concerned with the study of British plants. 222 THE JOURNAL OF BOTANY Mr. Magnus Spence's Flora Orcadensis contains much of interest. Dr. Irvine Fortescue tells us in his "Foreword" that it "is the result of many years of careful observation and research, and the author has spared neither time nor trouble in making his work as full as possible." The introductory portion includes among other interesting matter accounts of the author's excursions and the notes in the section entitled, not very happily, " Natural Selection " ; there is also useful biographical matter concerning previous contributors to our knowledge of the flora ; the sketch of the geology is very comprehensive. The list of species, in which introductions are prominent, suggests that Mr. Spence has not, j^C'Ce Dr. Fortescue as quoted, done quite all that he might have done to elucidate his subject, and there is an air of uncertainty about some of his entries which detracts a good deal from their value. Thus of a " var. montana " — a name with which we are not familiar — of Vicia sejnum he writes: "This is said to be found in Lyradale, Redland. I have not seen it, and do not know on whose authority it is given"; while, having definitely recorded Alchemilla vulgaris var. montana on his own authority, as well as on that of a contributor, he says in a later paragraph, " Rev. E. F. Linton, in a paper on the segregates of A. vulgaris [sic] , states that A. mon- tana is not British : in that case the variety in Orkney will be A. fiUcaulis," which has already had a separate entry. Of Eiyilobium tetragonum he says: — " Some doubt has been expressed as to whether this plant grows in Orkney ; but I believe it is to be found in several places in St. Andrews": surely this might have been decided ? The most interesting note in the book is that contributed by Mr. Moss (p. 138), in which he refers to a plant sent by Mr. Spence (who however had pointed out its differences) to Mr. Hunnybun as Primula scotica : of this Dr. Moss had at first regarded it as a variety, but he now considers it possibly identical with P. striata Fries, although further material is necessary before this can be decided. Those who use the book must not overlook the numerous additions to the text on pp. 127-137 : another appendix of " plants used medicinally " w^ould have been better incorporated in the text. So-called " English " names are given throughout : there are also a few local names of interest — Achillea Millefolium, " meal-an-folly " — a corruption of the specific name ; Artemisia vulgaris, " bulwands " and " grobbie " ; Spircea Ulmaria, " yule-girse " ; Menyanthes, " craw-shoe " ; seeds of Spergula, "reuth " ; Tormentilla, "hill-barks." There is a list of Mosses by Lieut.-Col. James Grant, of whom, as of Mr. Spence, a portrait is given ; with the exception of these and of the ferns and allies and Charas, only phanerogams are included. The book, interesting as it is, is one which might easily have been better had it been submitted to a competent " reader," who would at least have corrected the typographical blunders, which are sadly numerous — we note on one page (141) " Lepidopitra," ROOK-NOTES, NEWS, ETC. 223 " Sperganum," " palustra," and misplaced capitals: he would also have instructed the author in the art of making references. We are indebted to Mr. Bennett for some notes on the work, which we hope to print in an early number. BOOK-NOTES, NEWS, dc. Joseph Eeynolds Green, who died at Cambridge on June 3, was born at Stowmarket, Sufl'olk, on December 3, 1848. He went to Trinity College, Cambridge, in 1878, having previously taken his B.Sc. at London University ; he graduated as M.A. in 1888 and took his D.Sc. in 1891, and was elected a Scholar of his college. After taking this latter degree he worked in Michael Foster's laboratory on the enzymes contained in the seeds of plants ; he then undertook researches on fermentation, and in 1899 published a book on Soluble Ferments. In 1887 he became Px'ofessor of Botany to the Pharmaceutical Society, an office which he held for twenty years, and was Hartley Lecturer on Vegetable Physiology in Liverpool University. In 1892 he was elected Fellow of Downing College, and in the same year was president of the botanical section of the British Association, at whose meetings he was a constant attendant. In 1895 he pub- lished a Manual of Botany, in two volumes, and in 1905 an excellent Introduction to Vegetable Physiology ; in 1909 appeared his continuation of Sachs's History of Botany, dealing with the period 1860 — 1900. x\t the time of his death he had completed a work on the History of Botany in England, which will probably be published. He became a Fellow of the Linnean Society in 1889 and was elected F.E.S. in 1895. The Journal of Genetics for June (vol. iv. part 1) contains a long paper by G. H. Skull, of the Carnegie Institution of Washington, on " A Peculiar Negative Correlation in CEnothera Hybrids." As a result of his experiments the author criticizes somewhat severely some of the conclusions arrived at by Dr. R. B. Gates in his papers on the genus published in the Linnean Transactions and elsewhere ; the paper is illustrated by two plates. M. Wheldale and J. & LI. Bassett write " On a Sup- posed Synthesis of Anthocyanin " ; N. I. Varilov, of the Agricultural Higher School of Moscow, has a paper on " Immu- nity to Fungus Diseases as a Physiological Test in Genetics and Systematics, exemplified in Cereals," more especially in wheat and oats ; and H. M. Leake, Economic Botanist to the Indian Government, has a preliminary note " On the Factors controlling the Ginning per cent, of Indian Cottons." To the meeting of the Linnean Society on June 4 Professor H. H. W. Pearson contributed a paper, " Notes on the Mor- phology of certain Structures concerned in Reproduction in Gnetuvi" — an investigation of (1) androgynous and pseudo- androgynous spikes of Gnetum Gnemon ; (2) the young embryo sac of G. africanum. The spike which bears the male flowers occa- 224 THE JOURNAL OF BOTANY sionally shows complete female flowers in the lowest node or nodes. These flowers become pollinated. Their occurrence no doubt explains earlier references to the presence of fertile seeds on the "male" spikes. One of the complete female flowers in this position possessed four envelopes, the outermost of which bore an axillary bud. The two envelopes which stand between the outermost and the innermost are interpreted as being derived by duplication from the normal middle envelope. In its vascular supply and other characters, the axis of the " male " spike of G. Gnemon resembles that of G. scandens, and differs from those of the West African species. The microspore- mother-cell is 4-chambered. The pollen-grains in the micropyle contain either two or three apparently free nuclei. In the nucellus of the female flower the " pavement tissue " described by Coulter for G. Gnemon is formed also in G. africanum. Lotsy's account of the appearance of a cellular endosperm in the chalazal end of the sac of G. Gnemon before fertilization is true for G. africanum and confirmed for G. Gnemon. In G. africanum the nuclei of these cells are all formed by fusion, as in Welwitscliia. The fusing nuclei appear to be both morphologically and physiologically equivalent to the free gametes of the micropylar end of the sac. The primary endosperm is therefore homologous with that of Wehvitschia. Seasons for regarding it as morphologically different from the prothallus of Ephedra and the lower gymno- sperms are discussed. The latest number (xlii. 286, July 14) of the Journal of the Linnean Society (Botany) contains a monograph of the African species of Crotalaria, by Mr. E. G. Baker. For the purpose of his paper, Mr. Baker has visited the Herbaria at Paris, Brussels, and Berlin, and from these, in addition to the collections in the Kew and National Herbaria, he has obtained material which has brought up the number of African species to 309, of which 76 were previously undescribed. "Of these novelties, 9 belong to the section Simplicifoliae, 21 to the Sphoerocarpae, 1 to the Spinosae, 45 to Eucrobalaria." The paper, which is accompanied by six plates, bears testimony to the great strides that have been made in our knowledge of African botany since the pubHcation of the Flora Capensis (1861) and the Flora of Tropical Africa (1871) in which 106 and 24 species are respectively enumerated. The collection of drawings of South African flora and fauna made by Eobert Jacob Gordon, of which some account was given on p. 75, has been purchased by a committee in Holland, formed for the purpose, at the price of £1250. The drawings wall be placed in one of the State museums. The Beport of the Watso7i Botanical Exchange Club for 1912- 13 has reached us ; we hope to give some extracts from it in an early issue. The Editor wishes it to be understood that he does not concur in the views as to nomenclature expressed by Dr. Moss in his paper on Alsine (pp. 196-201). Journ.Bot . Tab. 533. w p. Highleydd.etlith. ' West, Nei«nan imp. Neosabicea Lehmannii Wern^haTrv. 225 NEW RUBIACEiE FKOM TROPICAL AMERICA.— IV. By H. F. Wernham, D.Sc, F.L.S. (Department of Botany, British Museum.) (Continued from vol. li., p. 324.) ^ (Plate 533.) Neosabicea Wernham, Rubiacearum e tribu Mussaendearum novum genus. Calycis lobi 4 subfoliacei plus minus angusti, elongati, sub- SBquales. Corolla subtubularis sub lobos parum ampliata ; lobi breves, valvati ; tubus intus insuper dense pilosa. Stamina 4, tubi basi inserta ; filamenta tenuissima elongata ; antherae dorso affixse, lineares obtusae. Discus inconspicuus. Ovarium biloculare ; stylus filamentosus inclusus apice bilobus ; ovula in loculis pauca complanata, placentis tumidis septo prope basin affixis adscenden- tibus plus minus impressa. Fructus . . . Herbge suffruticosse scandentes. Folia opposita ; stipulae minimae fere obsoletas. Flores parvi in racemis axillaribus dispositi. N. Lehmannii Wernham, sp. unic. F^rutex volubiHs, ramulis pubescentibus ; foliis ellipticis vix acuminatis obtusis basi non- nunquam cuneatis + 5 cm. x 3 cm., supra scabridulo-pubescen- tibus subtus pubescentibus, venis subtus prominulis secundariis utrinque ca. 8, petiolo brevi 2-5 mm. longo, stipulis fugaciis ; racemis paucifloris pubescentibus, ca. 5 cm. longis ; calycis lobis lanceolatis subacutis vix ad 5 mm. longis ; corolla extus sparsi- uscule pubescentis tubo insuper parum ampliata vix 1 cm. longo, lobis ovatis subacutis ca. 4 mm. x 3 mm. Colombia : Cauca, Popayan, 7150 ft., LeJmiann, 3514 ! Hb. Mus. Brit. Remijia Trianse Wernham, sp. no v. Frutex ramis crassis angulatis ferrugineo-villosis, novellis dense tomentosis ; foliis oppositis V. ternatim vei'ticillatis subcoriaceis ellipticis ad ca. 40 cm. X 18 cm. basi angustatis leniter acuminatis, supra glabratis subtus praesertim in venis hispidulo-villosis, venis supra impressis subtus prominulis secundariis utrinque ad + 20 ; petiolo valido basi nonnunquam tumido ad 4 cm. v. longiore ; stipulis maximis tarde deciduis foliaceis ovato-lanceolatis acuminatis acutis ad 8 cm. x 3 cm. extus rufo-hispidulis ; inflorescentia ad 35 cm. longa, bracteis submembranaceis stipularum forma multo tamen minoribus + 2 cm. X 8 mm. ; floribus parvis, calycis lobis brevissimis triangu- laribus, corollce, tubo + 8 mm. lobis subaequantibus; fructu ovoideo, ca. 2-5 cm. x 1*5 cm., glabrescente. Colombia : Villavicencio, Plain of San Martin, 1300 ft., Triana, 3273/1 ! Hb. Mus. Brit. Allied to the Brasilian B. ferruginea DC, but distinct, especi- ally in the much larger and differently-shaped leaves, with longer petioles, larger persistent stipules, short, broad calyx-teeth, &c. Declieuxia peruviana Wernham, sp. nov. Suffrutex ramulis virgatis novellis gracilibus fere glabris tardius cortice ai'genteo- JouRNAL OF Botany. — Vol. 52. [September, 1914.] s 226 THE JOURNAL OF BOTANY griseo indutis ; foliis linearibus subcoriaeeis + 1-3 cm. x 2 mm. obtusis sessilibus margine revolutis glabris venis nisi subtus centrali prominula occlusis, stipulia triangularibus ca. 2 mm. x 1 mm. conspicuis acutissimis rigidulis ; injlorescentia subcapitata l"5-2 cm. in diam. ; florihus parvis vix 5 mm. longis extus glabris. Peru : Chachapoyas, Mathews I Hb. Mus. Brit. Eemarkable for its small linear leaves and delicate habit. Declieuxia roraimensis Wernham, sp. nov. Glaberrimus, foliis lanceolatis 7-11 cm. x 2-3 cm., utrinque angustatis apice subacutis, petiolo brevi vel ad 1-3 cm. longo, stipulis lanceolatis acuminatis 7 mm. x 3 mm. infra medium bifidis ; florihus parvis in cymis corymbosis folia excedentibus longe- (5-6 cm.) peduncu- latis dispositis, calycis limbo subintegro, corollce tubo 5 mm. longo insuper infundibulari lobis oblongis obtusis 2-3 mm, longis. British Guiana : Roraima, Scliomhurgk, 581 ! Hb. Mus. Brit. Differs from all other species known to me in the habit, which closely resembles that of many Psychotrics, as also in the bifid stipules. De Candolle (Prodr. iv. 481) described a plant which he named suggestively D. ? i^sycliotrioides, expressing, however, a doubt as to the genus, as no flowers were present. This also has bifid stipules, according to the description ; but they are said to be subulate, the peduncles short, the inflorescence falling con- siderably short of the leaves. De Candolle gives as localities for D. 2^sychotr hides, Cayenne {Pair is), and Panama and Mexico {Haenke). LiNDENIA. The genus Lindenia is an interesting member of the tribe PhOndeletiea, with long-tubed, showy flowers. Two species have hitherto been known from the New World, L. aciitiflora and L. rivalis, both natives of Central America, and forming the respective subjects of plates 475, 476 in Hooker's Icones. The genus comprises two other species, L. austro-caledonica Brongn. and L. vitiensis Seem., the locality being indicated in each case by the specific name. The flower typical of Lindenia recalls that of the large- flowered MussandcB with subequal calyx-lobes. The two following species are very distinct from those previously described ; both are preserved in the Kew Herbarium, to the authorities of which I am greatly indebted for the privilege of examining the material. L. radicans Wernham, sp. nov. Frutex repens, caulibus gracilibus novellis appresse pubescentibus demum glabrescen- tibus ; foliis membranaceis ellipticis utrinque angustatis acutis acuminatis, 5-8 cm, x 2-3 cm., cystolithis breviter linearibus dense onustis, venis utrinque appresse hirtellis, secundariis utrinque 4-6, iMiolo gracillimo ad 2 cm. longo, stijMlis e basi brevissime triangulari subulato-setaceis ; floribus solitariis sessili- bus ; calycis laciniis linearibus acutis inaequalibus ad ca. 2 cm. longis ; corollce tubo gracili extus appresse pubescente ca. 4-5 cm. longo, lobis latis dorso subsericeis margine ciliatis rotundatis nee acuminatis ca. 1"5 x I'l cm., ore minute necnon densiuscule ALPINE VEGETATION ON BEN-Y-GLOE, PERTHSHIRE 227 ferrugineo-fui-furaceo ; staminibus omnino inclusis ; ovario dense appresse griseo-pubescente biloculari. Mexico : shaded places on rocks at crossing of the river near Tocotepeque, Jvily, Hartweg ! Hb. Kew. Remarkable for the creeping habit, thin leaves with densely packed and conspicuous cystoliths, rounded corolla-lobes, and included anthers. L. acuminatissima Wernham, sp. nov. Frutex, /o//is perga- maceis lanceolatis utrinque angustatis acutis ad ca. 3 x 1"2 cm. supra subnitentibus necnon glabrescentibus subtus pubescentibus, brevissime petiolatis, stipulis brevibus acutissimis rigide subu- lato-triangularibus. Floribus in axillis solitariis ; calycis lobis subulato-linearibus ca. 4-7 mm. longis ; corolla, tubo ad 7-8 cm. longo extus dense appresse sericeo, lobis 2-6 x 1 cm. ellipticis caudato-acuminatis acutissimis. Barclay ! in Hb. Kew. The label bears the names "Gardenia mitis. Tobago Jasmine " ; the exact locality is thus, unfortunately, doubtful. The new species is allied to L. rivalis, but it is readily distinguishable therefrom by the lengthy acumination of the corolla-lobes and the much shorter calyx-lobes. Explanation of Plate 533. — Neosahicea LehmannWWernh&m. — 1. Portion of flowering-shoot, natural size. 2. Single flower, x 4. 3. Flower in longi- tudinal section, x 4. ^^ ALPINE VEGETATION ON BEN-Y-GLOE, PERTHSHIRE. By Albert Wilson, F.L.S., and J. A. Wheldon, F.L.S. In the copious literature devoted in recent times to the subject of plant geography, the constituents of the chief plant associations of Great Britain have been well discussed, and the subject has reached a further stage in its evolution, viz. the investigation of the inter-relation of the various units included in each group, and the causes which determine the presence or absence of certain of them. This necessitates a careful biological study of each species, and an inquiry into its means of adaptation to the ecological conditions of the habitat. Although this advanced stage has been rapidly reached in the case of the higher plants, owing to wide floristic knowledge gleaned by generations of enthusiastic "field botanists," our information on the grouping of associations dominated by cryptogams is still incomplete, and the literature somewhat scanty. In the various associations dominated by spermopbytes, their cryptogamic associates have not been fully worked out, and only the more conspicuous ones are mentioned in lists of characteristic species, frequently, indeed, the generic name alone being deemed sufficient, e. g. Bryum, Sjjhagmini, Polytrichum, Cladonia, &c. Yet these genera contain species insisting on very varied climatic and edaphic conditions, and as a rule it will be s 2 228 THE JOURNAL OP BOTANY found that cryptogamic plants are very sensitive to such in- fluences, and often form a reUable index to the cHmatic and geological conditions of the environment. But before the effect of these conditions can be adequately gauged, there is much preliminary work of a systematic and floristic nature to be done. These few notes, meagre as they are, may therefore have their value as a small contribution to our knowledge of one of the more inaccessible and less easily worked " formations," viz. the Upper-Arctic Alpine Formation. Before going further it may be well to explain what is meant by the term Upper-Arctic Alpine Formation. It consists briefly of the flora of the higher mountain summits in Great Britain. Where the continuous undulating and rising ground begins to be differentiated into individual summits, a change from the pre- vailing moorland and subalpine vegetation becomes noticeable. In addition to the characteristic plants of such habitats, we begin to find examples of Arctic types ; and, to meet conditions of greater exposure, the morphological features of the species become modified in various ways, with which botanists are now familiar. At 2700 ft. and upwards, the Upper-Arctic Alpine Flora predominates. It is divided by Moss (1) into three groups. The first, the Chomovhyte Association of Crags and Carries, does not concern us here, as we did not meet with this class of habitat on the route by which we ascended Ben-y-Gloe. The two remaining divisions are sections of what K. Smith (2) termed the " Alpine plateau " and Warming " Fell-field." One of them, the Moss-Lichen Association, is an open association occurring on fairly level plateaux and scree-slopes, the vegetation being sparsely scattered over ground strewn with gravel, stones and boulders. The other is a closed association, occurring on slopes of usually less broken ground at similar altitudes, known as BJiacomitrium Heath, which at its lower limits, and under certain conditions, passes into various moorland groups. Two other associations are possibly to be separated within the formation — that dominated by Antlielia (3) on wet rocky slopes ; and the Marsupella Association of Macvicar (4), which is partial to slopes facing east and north, and is dominated by hepatics and such plants as Dicranum falcatnm. The two associations des- cribed in this paper are the Moss-Lichen Association and the Bhacomitrium Heath, which are well-marked and distinct on Ben-y-Gloe, although they frequently show transition stages, and become more or less intermingled. Such modifications we have observed on a slope of Braeriach towards Glen Eunach in West Invernesshire, and elsewhere. The flowering plants of these associations have been fairly well worked out and recorded, and we do not propose to devote attention to them now. Many of our most interesting Highland plants are included amongst them, such as alpine species of Potentilla, Alchemilla, Vaccinium, Arctostaphylos, Saxifraga, Azalea, Salix, Juncus, Luzula and Carex. The species are, ALPINE VEGETATION ON BEN-Y-GLOE, PERTHSHIRE 229 however, less numerous than in the Chomophyte Association, or in the groups of the lower slopes of the hills. The Cryptogamic Flora, on the contrary, requires much further exploration. In Mr. Tansley's Tyjjes of British Vegeta- tion, the flora of the Bkacomitrium Heath is represented by a list compiled from field notes by K. Smith and C. E. Moss, and from indications in local floras. The only cryptogams mentioned in this list are Bhacomitnum lanuginosuvi, B. ericoidcs, Cetraria islandica, Cladonia rangifcrina and Peltigera canina. It may be remarked in passing that, of the subsidiary species named, Cladina rangiferina was the only one we saw on Ben-y-Gloe. In the list representing the Moss-Lichen Association in the same work (1), no mosses or lichens are included, which is rather suggestive of a certain Shakespearian play produced with the part of Hamlet excised. But the florultB of these associations vary greatly on different mountains, and even on different portions of the same massif, according to the altitude, aspect, degree of moisture, steepness of slope, and rock-constituents. In a more recent work by C. B. Crampton (5), the plateaux debris of quartzite hills in Caithness is stated to contain the following mosses and lichens: — Bhacomitrium lanuginosum, Hyjmum Schreberi, and Cladina, spp. Dr. Moss says {loc. cit.): "The lists of species available at present do not warrant any attempt to draw up complete lists ; as to the lower plants there is little information." It was principally for this latter reason that, happening to meet with fine examples of these associations on Ben-y-Gloe, we deemed it advisable to make a few notes on their constituents. This mountain, " the mountain of the mist," attains a greater height and has a finer outline than any other in Perthshire, east of Tay and Garry. It is situated about seven miles north-east of Blair Atholl and rises direct from Glen Tilt. There are two peaks, a western one having an altitude of 3505 ft., and an eastern — known as Carn nan Gabhar or " Cairn Gowar " — rising to 3671 ft. We spent but a few hours on the mountain, and owing to thick mist only reached the lower of the two peaks. The ascent was made from Glen Tilt in rain and mist, on May 6th, 1912, the course being up the north-west side of the mountain. The aspect presents very little crag or rock exposure, and loses the interest provided where the chomophytic formation is present. Where this latter exists contiguous to the associations we are dealing with, the species are often reinforced from the crags and show greater variety. The lower slopes of Ben-y-Gloe are covered with the usual grass association so common in the Highlands, above which there is a considerable area of heather moor, with, in places, patches of ground dominated by Scirpus caspitosus. In the heather moor the chief Sphagna noted were compact tufts of Sphagnum ruhellum, S. fuscum, and a little S. subnitens, thus differing from our Lancashire moors, on which S. p>apillosum predominates, and S. fnsciim is practically absent. The ling thins out as the ground rises, and above 2500 ft. becomes much 230 THE JOURNAL OF BOTANY mixed with crowberry [EmiMrum nigrum) and some bilberry {Vaccinium Myrtillus). At about 2800 ft., near the shoulder of the flat-topped ridge or plateau which leads up to the summit on the south-west, the ground is somewhat broken and stony, the association becomes gradually more open, and eventually resolves itself into a fair area of the Moss-Lichen Association. We were able to make a careful examination of only a small portion of this ground, and the list of species submitted below is of necessity incomplete. The season was very early for this altitude, and the time allowed by train arrangements was all too short for search- ing out the more minute cryptogams. Moreover, the different classes of these plants require searching for on separate occasions. We have found from experience how easily mosses and hepatics may be overlooked when Hchens are being collected, and how incompatible the quest of flowering plants is with that of the smaller cryptogams. Subject to allowance for these limitations the following is a complete list of all the species noted on this detached portion of the Moss-Lichen Association, occurring on boulders, stones, and thin soil overlying the quartzite rock : — Flowering Plants. Alchemilla alpina L. Empetrum nigrum L. Gnaphalium supinum L. Vacciniitm Myrtillus L. Salix herbacea L. Aira and other grasses not in flower. Vascular Cryptogams. Lycopodium Selago L. L. alpinum L. Mosses. Andrecea petrophila Ehrh. Polytrichum alpinum L. P. piliferum Schreb. Dicranum fuscescens Turn. D. scoparium v. turfosum Milde. Bhacoviitriiim hcterostichum Brid. B. lanuginosum Brid. Webera nutans Hedw. Hepatics. DiplophyJhwi albicans Dum. Fungi. Ticothecium erraticum Massal. Lichens. Cladonia cervicornis Schaer. Thamnolia vermicnlaris Schaer. Lichens {continued). Stereocaulon coralloides Fr. S. evoliUum Graewe. Spharopkorns fragilis Ach. Cetraria aculcata Fr. Platysma triste Cromb. Gyroi^hora polyphylla T. & B. Parmclia alpicola Fr. fil. P. lanata Wallr. P. lanata var. reticulata Cromb. Lecanora polytropa Schaer. L. badia Ach. TIceviatomma ventosum Mass. Lecidea aglcea Sommerf. L. fuscoatra Ach. L. Kochiana Hepp. L. lithophila Ach. L. lapicida Fr. L. auriculata Th. Fr. L. contigiia Fr. L. confluens Ach. L. fusco-cinerea Nyl. L. limosa Ach. L. demissa Th. Fr. L. griseoatra Schaer. Buellia badioatra Koerb. var. atrobadia A. L. Sm. Bhizocarpon geograpliicum DC. B. geographicum DC. var. atro- virens Koerb. B. confervoides DC. ALPINE VEGETATION ON BEN-Y-GLOE, PERTHSHIRE 231 Ascending another 700 ft., the summit plateau is reached, consisting of tlie same kind of broken gravelly ground, with scattered stones, and patches of very thin soil, formed chiefly of the debris of cryptogamic plants. The flora at this elevation is slightly different. The dominant plants are Carex rigida and Alcliemilla alpina. The quartzite blocks, of which the cairn is built, yield addi- tional Gyrophorce, viz. G. crosa, G. torrefacta, and G. cylmdrica, the first and last-named being very scarce. On loose stones we saw a small quantity of a sterile Pyrenopsis, but Parmelia alpicola was fairly well developed on quartz crystals. On half-decayed patches of Cladonia, Bhacomitrium and Dicranum fmcesceus there is an abundance of Lccanora tartarea var. f rigida, with Lecidea arctica Somm. and L. limosa Ach., on peaty soil fiUing the crevices of the stones. These two lichens appear to thrive in the most bleak and exposed situations offered by our mountains. They are accom- panied by a few starved-looking examples of Thamnolia vermicu- laris, a curious Cladonia-like plant, resembling in shape a small white earthworm. So far as we observed, this Hchen is much less fine and abundant here than in the moss-lichen association of the granitic summits of the Cairngorm range. We saw here the hepatic Gymnomitrium obtusuvi, and no doubt other minute species might have been noted had conditions permitted a more careful search ; but the Bryophyta were not obtrusively evident. Two lichens, Lecidea tabidula Nyl., and L. deparcula Nyl., are recorded by Crombie (6) as occurring on small stones on this summit, but we failed to find them, although they doubtless enter into this association, as also does Lecidea nigroglomerata A. L. Sm. on Cairn Gowar. Turning now to ecological considerations, this particular ground, from the broken nature of the surface, which is weathered into stones and gravelly detritus, is subject to rapid drainage, and there is practically no available subsoil water for the plants near the surface. The hard crystalline rock is markedly dysgeogenous, and not retentive of moisture. Any soil formed is either washed down into the interstices, or completely removed by the fierce gales to which these plateaux are exposed. But for the frequent showers and cloud-fog, and occasional pockets in w^iich a little humus is detained by the matted stems of cryptogams, the higher plants would scarcely be able to exist. In such situations the amount of rainfall is of minor importance as compared with its frequency, and probably the plants depend more on the prevalent cloud-mist than on rain for their supplies. There are intervals of bright sunshine and strong wind, producing rapid evaporation of moisture. Added to this frequent desiccation there is consider- able fluctuation of temperature, producing together a set of con- ditions which reduce the possibilities of plant life almost to zero. The special contrivances by which certain flowering plants are enabled to face these untoward circumstances are well known and need not be repeated. As regards the Bryophyta, they are mostly 232 THE JOURNAL OF BOTANY small and densely tufted, and the tufts are frequently sunk in crevices of the rock or ground which they completely fill, allowing little purchase to the wind. The leaves are frequently very hygroscopic, as in chomophytic species, in wet weather open, in drought closely appressed and imhricate. The stunted growth, often attained by a shortening of the internodes, brings the leaves closer together, so that when appressed their apices alone are exposed. The cells are usually small, thick-walled, and often the thickness is increased externally by papillae. This seems to be especially the case with mosses having comparatively blunt leaves. These papillae probably act, like the apices of the leaves, as foci on which moisture condenses during cloud-fog. The Lichens, also, are usually diminutive in size, or, if of larger growth, more or less compact and cushion-like. The only Cladonia noted was a small form of C. cervicornis, in which the thallus was reduced to a dense squamulose cushion, and the podetia very diminutive. Stereocaulon and Spliaropliorus occur in very compact forms, the outer stems prostrate, the next inclined, the central ones erect, so that the points only are exposed, and no lateral stress is received in wind-storms. The only foliose Parmclia seen was the closely appressed P. alpicola, which is so reduced as to resemble a crustose Lecanora. The laciniae of the thallus are very convex, imbricate and complicate, and often torulose, so that when wetted the water is partly retained in the very numerous depressions thus formed in the thallus, and such as finds its way under the subtubular divisions is retained. The tough leathery Gyrophor(B are centrally affixed, and often depressed at the margins, forming shields beneath which moisture is retained longer than on the open rock. In the case of G. torrefacta the under side is often fibrillose and covered with several trabeculate membranes, the whole producing a sponge-like texture. Occasional perforations admit moisture under the thallus. Lecanora tartarea assumes a very different appearance from that which it presents on trees or rocks at lower altitudes. It creeps over mosses and plant-roots, giving off at intervals slender spinulose processes, w'hich no doubt act as dew-collecting points. The erect fruticose lichens, so plentiful in the next group, are almost absent. They are repre- sented by Parmelia lanata, a decumbent or prostrate appressed plant with the appearance of Alcctoria nigricans, but a totally different habitat and mode of growth ; and Thamnolia vcrmicu- laris, a prostrate plant which hardly leaves the ground except where it turns up its pointed ends as an attraction to the dew. Many of the Lccidece. have large fruits, and very little thallus. The hyphae ramify in minute interstices of the stone, and the gonidia cluster under the lee of their own apothecia, so that they often at first sight appear to be quite athalline. This is especially the case on loose stones of small size, where the conditions are intensely dry. Like the scouts of an army, these outposts of vegetable life lead a precarious existence, and they take " cover " behind any prominent object, especially on the leeward side. Their growth ALPINE VEGETATION ON BEN-Y-GLOE, PERTHSHIRE 233 in a horizontal direction is often determined by the height of some neighbouring tuft or stone, in the shelter of which they lie. The attitude of many of the species may be described as one of " crouching " to obtain shelter from the wind. When shelter and moisture are both denied them, they have still a defence in reserve. Before a prolonged drought they simply adopt a con- dition of suspended animation. How long they can so live would be difficult to ascertain, but they must certainly be able to lie dormant for several weeks without injury. From the lower to the higher of the two plateaux or terraces described above, the ground rises in a moderate slope, and affords an excellent example of the closed Bhacomitrium Heath Suh- association. The requisite conditions appear to be rapid drainage, shallow soil, and a low mean temperature. Another condition which has been suggested as favourable for the production of Bhacomitrium heath is the absence of direct sunshine from northern slopes for several months during the winter, ov;ing to the low angle of elevation of the sun. The ground is also usually too steep and wind-swept for great accumulations of snow. On Ben-y-Gloe the drifts or Schneefiechten which we encountered lay below the Bhacomitrium heath. Except where the thick dense carpet of Bhacomitrium has been accidentally torn away, or where jutting rocks protrude through its compact layers, this association in its most perfect development contains few or none of the higher plants. The stems of the moss become procumbent, overlapping each other to a considerable depth, and the subsidiary vegetation consists mainly of lichens attached to its decaying branches. Macvicar (4) in alluding to the Bhacomitriiim heath says " the hepatics are almost absent from it," and so we found it here. We are able also to agree fully with the following statement by Moss (1) : " While the closed Bhacomitrium association would appear to increase the shelter for other species, it is noteworthy that the proportion of Highland species is generally less in it than on the more open stony waste." It is probable these remarks were intended to apply to flowering plants, but they are also applicable to the mosses, hepatics, and lichens. Tlie Bhacomitrium heath forms a delightfully soft and springy carpet to the feet of the traveller, and the change from the arduous toil of the heathery lower slopes is always welcomed by the climber. We found the flora here, as on similar tracts of other mountains, poor and scanty. It is doubtful if a prolonged search even at a more favourable time of the year would have added very materially to the list of species noted in the centre of the moss-carpet, i. c. where the association is closed and perfect. Probably, however, a considerable number of small cryptogams and encroaching spermophytes might be detected at various points where it merges into other formations. The paucity of species in the closed Bhacomitrium heath overlying quartzite rocks will be seen from the following list, in which the species are arranged in order of frequency. 234 THE JOURNAL OF BOTANY Dominant. Lycoj^odium alpinnm L. Bhacoviitrium lamujinosum Polytriclium aljnnum L. Brid. Cetraria aculeata Fr. Abundant. Cladonia cervicornis Schaer. Cladina rangiferina Nyl. ^- 0 r acil i s B.o&n^ Cetraria crispa Nyl. C. macilenta Hoffm. v. coronata Alectoria nigricans Nyl. ^ ' Eare. Frequent. Cladonia destricta Nyl. Cladina sylvatica Nyl. Lecanora tartarea Ach. C. uncialis Nyl. Cladonia hellidiflora Floerke. C. uncialis Nyl. v. ohtusata Nyl. C. squamosa Hoffm. forma. C. uncialis Nyl. v. turgesccns G- degencrans Floerke. Cromb. G- degcnerans Floerke v. plcolc- Cladonia deformis Hoff'm. pidea Nyl. Empctriim nigrum L. Alectoria ochroleuca Nyl. BcBomyces ariiginosus DC. Occasional. Bilimhia melana Arnold. Hypnum Schreberi Willd. Hjfpnum cuprcssifornie L. Ckf/07Ma furcata Hoffm. ^7/ ocom4W7;Worcm^t B. & S. Dicramim fuscescens Turn. PtiUdmm ciliarc Hampe. Carea; n^eVZa Good. ^^^^^^'"^ hP^orum Fr. The majority of the species in this list ascend considerably above the 3000 ft. contour. The vegetation of the plateau gravel has been described as cushion-like or crustaceous ; that of the present association may be said to be mat- or carpet-like. The dominant moss has long, much divided, trailing stems, which are interlaced together by numerous short hooked branches and long leaves. This clinging of the stems does not always prevent strong gales from tearing up large sheets of the carpet. Bare patches formed in this manner revert to the moss-lichen stage until the Bhacomitrium reassumes dominance. The apices of the leaves of this moss are devoid of chlorophyll, papillose and eroso-ciliate, thus presenting innumerable small prominences for the reception of dew. It will be observed that this echlorophyllose area is as papillose as the rest of the leaf, and therefore we cannot in this instance regard the papillae as simply designed to shield the chlorophyll from the sun, or to prevent evaporation, which is usually accepted as their raison d'etre. The condensed moisture, or, during show^ers, rain-water, runs down to the thick underlying mass of old stems and leaves, whicla hold it like a sponge. Still lower a layer of humus is rapidly formed. Most of the flowering plants and other dependent species cower beneath the shelter of the moss, only in favourable seasons protruding their growing tips, and, as is apparent from the paucity of individuals as well as species, often suffering severely for their temerity in so doing. The smaller mosses and lichens, especially those of effuse hori- zontal growth, have no chance, and are rapidly smothered, the erect fruticulose lichens greatly preponderating. Cladina and CladonicB with elongate podetia are frequent, presenting often ALPINE VEGETATION ON BEN-Y-GLOE, PERTHSHIRE 235 spinulose points al)Ove the moss. Cciraria crispa straggles up through the mats, singly or in dense tufts. Its margins are beset with cilia, giving it a distant resemblance, on a larger scale, to the eroded leaf-margin of the dominant moss. In dry weather the thallus rolls up and becomes subtubular. Some of the large- branched Cladonice have perforate axils, the orifices being at times spinulose, the spines acting as water conductors to the interior of the hollow podetia. A lichen almost confined to this association is Alectoria nigricans, and the allied A. ocliroleuca is also partial to it. They resemble miniature trees. The trunks obtain support by standing buried in the Ehacomitrium, the branches rising above the surface, where they ultimately become dark and discoloured by exposure to sun, wind, and frost. In dry weather they are readily overlooked, but under moister conditions their entangled ramuli and slender branchlets become everywhere studied with tiny drops of dew, and are then very noticeable. When dry they are rigid and brittle ; then the feet of the alpine hare or ptarmigan readily crush them into small particles, to be disseminated by the wind. In this manner this rare lichen is probably propagated ; its fruit is unknown with us, but has been found, according to Crombie (7), in Labrador and Arctic North America. It must be observed that this list of species applies only to the Upper Arctic-Alpine formation of Ben-y-Gloe, and that had we been dealing with the similar formation of the micaceous debris of Ben Lawers, or granitic detritus of the Cairngorms, we should have a very different list of species, and in the case of the former a very much richer one. Such beautiful plants as Solorina crocea on Lawers, and Platysma nivalc on Braeriach, with many others as rare if not so handsome, would reward the explorer of the Arctic-x\lpine area of those mountains. A comparative list of plants from the various summits above 3000 ft. would provide exceedingly interesting matter for speculation. References. (1) A. G. Tansley and G. E. Moss : Types of British Vegeta- tion. 1911. (2) E. Smith : Botanical Survey of Scotland. Scott. Geog. Mag., 1904. (3) W. G. Smith : Anthelia, an Arctic- Alpine Association. Scott. Bot. Keview, No. 2, April, 1912. (4) S. M. Macvicar: The Distribution of Hepatics in Scotland. Trans. Bot. Soc. Edin. xxv., 1910. (5) C. B. Crampton : The Vegetation of Caithness considered in Eolation to the Geology. 1911. (6) J. Crombie: Grevillea, i. p. 62, and viii. p. 112. (7) J. Crombie: Monograph of British Lichens, part i., 1894. 23G THKEE CONIFERS. By R. a. DiJMMER. Thuya (Biota) orientalis var. nov. mexicana Diimmer. In 1817 Humboldt, Bonpland, and Kunth (Nov. Gen. et Sp. ii. 3) des- cribed Ciipressus thurifera from specimens collected at Tasco and Tehuilotepec in Mexico ; it was subsequently detected by Uhde in Michoacoa and Oaxaca, and on the Sierra Madre by Seemann, and Hartweg averred having seen specimens of it 120 ft. high near Real de Monte ; since that time it has remained undiscovered, and, as the paucity of herbaria material suggests, is one of the rarest of conifers. Thirty years later Endlicher (Syn. Conif. 62) included this species in the genus ChamcBcyimris, whence the name Chamoi- cyixtris thurifera; but Masters having examined fragments of the original specimens, which are preserved in the Willdenow Her- barium, Berlin, and at Paris, restored it to its original position. In his critical account of the genus Cupressus, Masters (Journ. Linn. Soc. (Bot.) xxxi. 349) gives figures depicting leafy twigs and cones " of this plant, which in no way agree with Cwpressus lusitanica, its variety Benthami, two cypresses with which it has been confounded. While engaged on the conifers of the Lindley Herbarium, Cambridge, my attention was drawn to a small fruiting specimen labelled " Chamacyparis thurifera Endl. 308, Culta. Sept. 5, 81. Orizaba," which suggested this long-lost plant. This particular specimen was described by Lindley himself as such in Gard. Chron. in 1856 (i. 772), and as his remarks are of interest, I here transcribe them : — " There is commonly found in gardens throughout Europe a Mexican coniferous tree called Cupressus thurifera, under the supposition that it is the plant so called by Humboldt. Endlicher, however, pointed out the mistake, showing that the garden plant is a true Cypress, while that of Humboldt is a Chamcecyparis, and therefore he called the latter Chamcecyparis thurifera, giving the name of Cupressus Benthami to the wrong-named garden plant. Dr. Klotzsch had previously circulated the name of C. Lindlcyi for the same thing. No two plants can be more different than Cupressus Benthami and Chamacyparis thurifera — the first a true Cypress with numerous seeds to each scale of the cone — the second a Chamacyparis with only two or one, and those not winged. " But nobody seems of late years to have met with this Chamoicyparis in Mexico ; even in the vast Herbarium of Kew it is not to be found. A few cones of a Cypress-like plant with roundish wingless seeds having, however, been received by the Horticultural Society, and raised in the Chiswick Gardens, further * The cones (fig. 27) are those of Thuya (Biota) orientalis var. mexicana — Botteri's cones, which are preserved at Kew and at Cambridge, and which Dr. Masters erroneously suggested might belong to this species. THREE CONIFERS 237 inquiry has led to the discovery that this Orizaba plant is in all probability the long-lost Chamaciji^aris thurifera. Botteri sent them home without one word of information, but with a small dried specimen, from which we learn that he found it in a culti- vated state. Being a true Gliamacijparis, and only one from Mexico being known to botanists, the inference that we have at last the true plant of Humboldt seems inevitable. In that great philosopher's work, above quoted, it is said to be a very tall resinous tree with spreading branches, whose timber is used for building purposes. The young seedlings in the Garden of the Horticultural Society are very glaucous, with almost the aspect of a Thuya. How far they may be able to bear this climate remains to be ascertained." Now, as the wingless character of the seeds of Lindley's specimen, in conjunction with the non-peltate nature of the cone- scales, excludes it from Gupressiis and ChamcBcyixiris respectively, it must be a Tlmya, and one which, though agreeing with C. (Biota) orientalls in its foliar characters, merits, on account of its peculiar cones and seeds, at least, in view of its reputed garden origin, varietal rank. It might have been premised that these fruit distinctions would be ultimately correlated with peculiarities in the disposition or shape of the leaves, but such is not the case ; the latter have remained stable, and show no salient points of distinction from Thuya orientalis, a plant of great horticultural age, which no doubt had even penetrated into so little known a region as Orizaba in Mexico, where possibly the change in edaphic and climatic factors evoked this mutant. Its cones are solitary, subsessile, light brown in colour, and subglobose and ^^ in. across, are com- posed of six non-peltate decussately arranged scales, which are slightly fused at the base, the larger outer being suborbicular, obovate or subrhombic in general outline, |-y\ in. long, ^'W-i in. broad, with a small scarcely perceptible dorsal process, and" slight longitudinal depression below it, their inner faces being marked at the base by the scars of attachment of one to three seeds ; the inner scales are not narrower, narrowly obpyramidate and quad- rangular, their apices flattish and slanting, with a small median process. The seeds are wingless, greyish brown and smooth, 4-6 mm. long, obliquely ovoid, obscurely 3-4 angled, bevelled on one side at the base, and opposite to this a semicircular lighter coloured scar. Hence the cones differ from typical Thuya orientalis in their smaller more globose shape, their light brown colour (no glaucous sheen being apparent), the different confor- mation of the scales, the absence of the horn-like dorsal recurved processes, so conspicuous a feature of the type, and the much smaller seeds, which in Thuya orientalis measure up to 8 mm. in length, and do not possess the peculiar basal bevelling charac- terising the latter. _ The plants raised in the Chiswick Gardens from the seeds of this variety doubtless perished, for no Thuya bearing such cones has ever been alluded to in horticultural literature, as far as the 238 THE JOURNAL OP BOTANY writer is aware, but there is reason to assume that the variety existed on the Continent, for Carrifere opines that the plant cultivated there as Cupressus thurifera was a Biota. Kent's description (Veitch's Man. Conif. 230 (1900)), based upon speci- mens received from La Mortola, apparently coincides with this rare species ; but Mr. A. Berger, in a letter to Messrs. Henry & Elwes, states that this plant has disappeared and that two plants formerly cultivated under this name at La Mortola turned out to be Cu2)ressus sevipervirens and C. Uisitanica var. BentJiami (C. Benthami) respectively; similarly, seeds of C. thurifera distributed in 1909 by the Dendrological Society of France differed in no respect from those of C. lusitanica. Quite recently Mr. Clinton Baker has received specimens from Mons. J. Daveau of the Jardin des Plantes, Montpellier, under the name of Cnpressus thurifera, which were collected in a cemetery of the State of Puebla, Mexico, by J. Nicolas (No. 9776) in January, 1911, but which Dr. Henry agreed with me must be assigned to Cupressiis (Biota) orientalis. The latter specimens have therefore confirmed our suspicions as to the cultivated nature of the species in that region. Two other cone-bearing specimens in the Lindley collection labelled C. thurifera Schlecht., collected between Angauguco and Italpuxahua and Banco, are referable to Ciipressus Uisitanica var. Benthami. Callitris neo-caledonica, sp. nov. The vegetation of the island of New Caledonia has been discussed by the French botanist Brongniart, by Professor Engler, and latterly in an able manner by Dr. E. Schlechter in Engler's Bot. Jahrbuch, xxxvi. (1905), who bases his observations upon several months' sojourn in the island. His sketch was subsequently followed by an enumeration and description [op. cit. xxxix. 1906) of the plants collected, among which many novelties occur; and the author lays special stress upon the advisability of exploring the north-west corner of the island, and the near-lying New Hebrides, regarding the flora of which our knowledge is regrettably deficient. In view of the remarks which follow, it appears expedient to allude to Schlechter's remarks respecting the Conifers wdiich obtain on the island. These are invariably confined to upper mountain-slopes between elevations of 1400-1500 metres above sea-level, only one, Agathis ovata Warburg, apparently affecting an altitudinal range from sea-level to 1300 metres, where, as solitary specimens, it is particularly abundant in the Serpentine Eegion. The Araucarias, of which A. Balanscs, A. montana, and A. Muelleri are cited, are among the tallest of the Conifers, rarely exceeding 35 ft. in height, and, being easily recognized by their distinct pyramidal aspect, form, in contradistinction to the remaining gymnospermous vegetation, small and nearly pure strands, especially in the south of the island ; Dacrydium is repre- sented by three species, D. Balanscs and the Yew-like and Arau- caria-iike D. taxoides and D. araucarioides respectively, which are scattered among woods of an Australian-Malayan element, chiefly composed of various Myoporums, Cunonias, Spiraeanthema, THREE CONIFERS 239 Soulameas, Eugenias, and several Myrtles, &c. Podocmyiis gnidioides remains dwarf, covers the ground, and in its prostrate and creeping habit mimics Juniperus Sahina; of other Podocarps, P. minor and P. usta, with broad spreading crowns, rarely exceed, like the Dacrydia and species of Callitris, 25 ft. in height. The latter are associated with the Podocarps and Dacrydia in the upper woods of the southern region of the island, and in exceed- ingly exposed situations remain dwarfed and shrubby. Libocedrus austro-caledonicus (L. neo-caledonicus of Schlechter) is seemingly rare ; it was detected by Schlechter in the woods on the slopes of Mount Humboldt at elevations of 5300 ft. Three species of Callitris are cited, G. Balansce, C. sulcata, C. suhumhellata ; the two latter were first referred to by Parlatore in the list of plants of the Florence Garden, and were subsequently described by him in De Candolle's Prodromus under the name of Frenela, where he cited specimens in Hooker's Herbarium. A reference to these shows that F. sulcata is based upon a fruiting specimen bearing scale-leaves, collected by Moore in New Caledonia, and communi- cated to Kew by Messrs. Veitch in August, 1862. F. suhumhellata is founded upon specimens which were obtained on dry stony ground at the base of the mountain-ranges south-west of New Caledonia by an unnamed collector, showing both scale and primordial acicular leaves, the latter free and arranged in fours, the former in threes, and like those of Junipers fused. These specimens can in no way be regarded as specifically distinct from F. BalanscB, and agree with Balansa's specimens of F. Balansce. of Brongniart & Gris., which also exhibit both foliar phases and where their intimacy is remarkably well shown. The synonymy is as follows : — Callitris sulcata Schlechter in Engler's Bot. Jahrb. xxxix. IG (1906) ; C. suhumhellata and C. Balansce Schlechter, I. c. Frenela sulcata and F. suhumhellata Parlatore, Enum. Sem. Hort. Florent. (1862), 23 ; in De Candolle Prod. xvi. ii. 446-447 (1868) ; Frenela Balansce Brongniart and Gris. in Bull. Soc. Bot. Fr. xvi. (1869), 327. Schlechter collected specimens of a Callitris on the mountains of Ngoye at elevations of about 3000 ft. in December, 1902, which he regarded as identical with G. Balansce, but a critical examina- tion precludes this and warrants their retention under a separate name; the salient points which distinguish this species from its congener lie in the short and moreover congested nature of the articulated twigs, the much shorter leaves with prominent dorsal convexities, and their denticulate not deeply and closely laciniate hyaline margins. The following is a description : — Callitris neo-caledonica sp. nov. A tree about 25 ft. high with a broad crown ; third year's branchlets stout, terete, rough, greyish brown, invested with the remains of the decurrent scale- leaves, current year's twigs short, articulated, 3-angled, leafy, 1-3 in. long, 1-7-2 mm. broad, ascending, fastigiate, densely crowded. Leaves (only homomorphic as far as is known) imbri- cate, scale-like, ternate, fused except their incurved triangular 240 THE JOURNAL OF BOTANY apices, averaging 3-5 mm. long, their dorsal convexities separated by shallow grooves ; apices free, triangular, acute, their margin not hyaline but opaque and microscopically denticulate. C. sulcatcB Schlechter affinis sed ramulis hornotinis articulatis brevibus valde confertis, foliis brevioribus, marginibus baud hyalinis vel laciniatis sed opacis minute denticulatisque diiifert. (Schlechter, 15179, Herb. Kew.) ! ^ PoDOCARPUS MoTLEYi Diimmer (comb. nov.). In 1857-8 Mr. James Motley collected specimens (No. 1300) of a Conifer near Bangarmassing (spelt also Bangermasin or Bandgermasin) in Southern Borneo, which he described as a large lofty tree with smooth yew-like bark, bearing the native name, " Kaju saribu dauni," meaning literally the " tree of a thousand leaves," in reference doubtless to its densely leafy crown. Some of these specimens fell into the hands of the celebrated Italian botanist Parlatore, who described and named the species Dainmara Motley i (without, however, seeing fertile material), in compliment to its discoverer. Parlatore's original description occurs in his List of Seeds of the Florence Botanical Gardens for 1862 (published 1863), and synchronously, if not earlier in this Journal for Feb. 1863 (p. 36) : Seemann [1. c.) in a footnote to Parlatore's description writes : "The genus must be considered doubtful, as the fruit is unknown. It may be a Podocarpus. Some time ago I asked the question in the Gardeners' Chronicle, how the Nageia section of Podocarpus could be distinguished from the genus Dammara in habit ; and since then Mr. Charles Moore of Sydney has drawn my attention to the fact that the Dammara is leafy, even after the branches are several feet long, whilst in Podocarpiis it becomes bare at a very early stage ; and, as far as I have been able to observe, this distinction holds good." This difference is decidedly apparent in specimens under cultivation : in the Temperate House at Kew the leaves are known to persist on the trunks of the various species cultivated there for 5-20 years ; moreover, another constant obvious vege- tative character separating these two genera appears to lie in the terminal buds ; which in Podocarpus are narrowed and always pointed, while in Agathis {Dammara) they are without exception broad and hemispheric or depressedly hemispheric. Messrs. Seward & Ford in their interesting account of the Arancarieae, recent and extinct (Phil. Trans, cxcviii. 317 (1906)), confirm Seemann's suspicions, having made an anatomical in- vestigation of the leaves ; they say " the lamina, which is almost isobilateral, is characterised by the occurrence of resin canals between [below] the vascular bundles, and by numerous thick- walled fibres, with hardly any lumen below the epidermis. In this and other features the anatomy conforms to that of leaves of Podocarpus Nageia." Prof. Seward in a letter adds that in Podocarpus the resin canals of the leaves are below, in those of Agathis between the veins. Taking these views therefore into consideration, there can be but little doubt as to the correct status THREE CONIFERS 241 oi Acjathis Motleyi; and this was strikingly shown when in 1868 Parlatore (in DC. Prodr. xvi. 2, 508) described fertile material, which he did not recognise as of the same plant — collected l)y Beccari (No. 2649) between 1865 and 1868 in Sarawak, North- west Borneo — as Podocaiyus Bcccarii. The following is a brief description of the plant, to which I have added its synonymy : — Current year's twigs short, sub- verticillate, distinctly pulvinate and narrowly ridged, terminated by narrow sharply acuminate buds, enclosed by 2-4 oval acuminately cuspidate denticulate-lacerate scales. Leaves oppo- site, ascending and hence overlapping, oval, acute, or shortly and sharply cuspidate, attenuate or rarely rounded basally, with a very short broad not twisted petiolar base ; 1-1| in. long, ^-f in. broad, coriaceous and rigid, doubtless dark lustrous green in the living state, indistinctly longitudinally striate when dried ; margin not recurved. Male strobiles unknown. Females flowers disposed singly in the axils of the leaves, borne on a short, stout, j^ in. long peduncle. Receptacles thickly fleshy, cylindric, damson- coloured. Seed globose, smooth and brownish with a slight glaucescent sheen, f in. in diameter. PoDOCABPUS Motleyi, comb. nov. Dammara Motleyi Parlatore, Index Sem. Hort. Bot. Florent. 26 (1862) ; in Seemann, Journal of Botany, i. 36 (1863) ; and in De Candolle, Prod. xvi. ii. 377 (1868). Agathis Motleyi Warburg, Monsunia, i. 185 (1900). Podocaiyus Beccarii Parlatore, ojj, cit. 508 ; Pilger, Taxacete, 59 (1903). Podocarpus sp., Seward & Ford, in Phil. Trans, cxcviii. 317 (1906), with figure of leaf. Nageia Beccarii Gordon, Pinetum, 186 (1875). <^ JOSEPH ANTHONY MARTINDALE. (1837-1914.) Joseph Anthony Martindale, who passed away in his 77th year on April 3rd, was one of our ablest British Lichenologists, and was recognised as such on the Continent as well as in this country. He was born on July 19th, 1837, at Stanhope, in the Weardale Valley, Durham. His father moved soon afterwards to Durham, and became first mathematical master at Bede College, remaining there, however, only a short time, for when young Martindale was only eleven years old his father was conducting a private school at Sunderland, and lecturing and writing on agriculture and chemistry. Joseph, who was the eldest of seven children, at that age obtained a medal for chemistry, amongst youths of eighteen and nineteen, under the examination of a well- known professor, but his father, with a stern rectitude, forbade him to accept the medal, an act of probity which Martindale himself in later years used to refer to as rather hard upon him. Journal of Botany. — Vol. 52. [September, 1914.] t 242 THE JOURNAL OF BOTANY On his father's death, which occurred when Joseph was thirteen years pld, he became a pupil teacher, was trained at the Battersea Training College, and was appointed to a school at Stanwix, near Carlisle, in 1857. On October 3rd, 1859, he came to Staveley as headmaster, an appointment which he held with great success until his retirement in 1902 ; after his retirement he continued to lecture under the County Education authorities. He was twice married ; to Mary Ann Seed in 1861, and to Emily J. Euthven in 1894, leaving six children by the first, and one by the second marriage. His eldest son, Mr. G. E. Martindale, inherits his father's botanical tastes. Somewhat reserved with strangers, but of a kindly and un- selfish disposition, Martindale took an active interest in politics and in the local management of the village, serving on the parish council and other bodies, acting as organist of the parish church, and joining in the Volunteer movement of 1878. Physically he was active and vigorous, making all his jour- neys on foot when acting as inspector of religious instruction for the council schools of Westmoreland. Intellectually he was a man of considerable ability and determination, doing with the utmost thoroughness and precision everything he took in hand. His hchenological studies led him to acquire a mastery of the German language after he was forty years of age. French he knew well, and was thus able to correspond in their own languages with Arnold and Nylander. Besides being a classical scholar, Martindale was famihar with Anglo-Saxon, and was versed in the Norwegian and Icelandic languages. He held strongly to Anglo-Saxon associations, and challenged the ultra-Norwegian theories held by some of the Westmoreland antiquaries, by material derived from local place- names. He was mainly instrumental in the discovery of an ancient British settlement at Millrigg, Kentmere, and in 1900 read a paper on the subject before the Cumberland and West- moreland Archgeological and Antiquarian Society. His interest extended to entomology, geology, and osteology, and he was an old and honoured member of the Kendal Literary and Scientific Society and a member of its Council from 1903-1913. Although best known outside his adopted county as a lichenologist, Martindale was a good all-round botanist. When he first took up the study of the botany of Westmoreland, he, with his usual thoroughness, collected all the records of plants of the period before Linnaeus, from 1597 to 1774, availing himself largely of Mr. Harry Arnold's rich library at Arnbarrow: these he found to number 153 species. He then followed up the labours of Thomas Lawson in 1638, the Quaker schoolmaster of Great Strickland and father of Lakeland botany, who sent to his contemporary, John Eay, a list of 150 local plants, and of the stations in which they grew ; and brought the records up to date by consulting those of Wilson and Hudson in the 18th, and Gough in the 19th century. Of the total number of plants, 1858, enumerated in the London Catalogue (8th edition), Mr. Martindale JOSEPH ANTHONY MARTINDALE 243 found records for 1023 in Westmoreland and Furness ! But ho was too conscientious to allow aliens and garden escapes to be recorded as natives, and thus reduced the number to 897 un- doubted native species. In order to work out their distribution, he coloured the local map into six river basins, viz. the Leven and Duddon, the Kent, Lune, Eamont, Eden, and Tees, and the map was published by Bartholomew. By the help of local botanists, he was able to give, besides his own list of 500 lichens and 138 fungi, a list of 360 mosses and 118 hepatics, besides algae, diatoms and desmids, and brought the results before a local Natural History Society in 1888. His own herbarium contained about 2000 flowering and about 1000 flowerless plants. Martindale appears to have begun the study of lichens about the year 1867, judging from a letter received from him February 25th, 1869, accompanying a series of north country flowering plants which he kindly sent for my herbarium, in which he says : " I have for the last two years done next to nothing among the phaenogams, all of ray spare time being fully taken up with the T 2 244 THE JOURNAL OF BOTANY study of lichens, and I find that I make but very Httle progress with them. On looking over my collection I am astonished at the great number which I have determined, to which the mark of ' doubtful ' is attached. Those I am certain of are very few in comparison." During the next twenty years he evidently continued the study of the group until he mastered them, publishing papers on the Eeindeer Lichen, and on the lichens of the Placodium murorum group, which showed a masterly grasp of the subject ; as well as a list of the lichens of Westmoreland, in the Naturalist for 1886-87 : this included many rare species, and several new to Great Britain. Among these latter were -.—Exihcheia Martinclalei Cromb., Collema isidioides Nyl. (Warton Crag, Cumberland), Collemopsls ohlongans Nyl., Calicium roscidum Fkh., Parmelia isidiotyla Nyl., Gyropliora sjJodochroa Ach., Lecanora flavocitrina Nyl., Lecidca acutula Nyl., L. decUnascens Nyl., Platycjrapha pcriclca Nyl. The progress of the list was arrested by the death of his first wife, which affected his own health ; it will, it is hoped, appear in a complete form in the botanical section for the county (which, at my suggestion, Martindale was engaged to undertake) in the Victoria History of the Counties of England. My first acquaintance with Martindale came through J. M. Barnes (1814-90) of Levens, Milnthorpe, a most genial and liberal correspondent, who in 1867 sent me Westmoreland mosses in exchange for those of Devon. This excellent bryologist told me that he, Martindale, and George Stabler (1839-1910) used to meet once a month at each other's houses, and then go out on exploring expeditions. This little group of botanists did much for the botany of the county, and their names are perpetuated in plants they discovered in the course of their work : Barnes, in Bryum B arnes ii \\ood; Stabler, in Anthroceros Stableri Steph., Marsupella Stableri Sj^vuce, and Plagiochila Stableri 'Pesirson; and Martindale in E])hebeia Martindalei Cromb. Like the majority of practical lichenologists he was not a believer in the Schwendenerian theory. In a letter to me on February 20th, 1912, he writes: "The Schwendenerian theory creates more difliculties than it seems to solve. It is passing strange that lichen gonidia should so closely resemble algae, but it would be much stranger that Palmellacecs should remain for untold generations in an initial stage, without going on to com- plete their cycle or without dying away. This must be the case, if Schwendener is right, with many imprisoned ' algie ' in the thallus of lichens, that have never been known to fruit, and have therefore never imprisoned any algae, since the original ger- mination of the spores from which they came. There are several other things altogether independent of the question of gonidia, the chief of which is that the fertilisation is not effected as in the Ascomycetes, that is, if we accept as correct the statements of fungologists respecting them. I have myself microscopically examined thousands of apothecia, and scarcely ever limited my work to looking at and measuring the spores, but took in the SrARTINA TOWNSENDII 245 whole organ. I have examined them in their earliest beginnings and there is nothing resembling a poUinodium. The fungi them- selves are degenerate plants descended from some chlorophyllous parentage, and my belief is that lichens and ascomycetes descend from some common ancestor, but have diverged just as man and the ape have diverged in different directions from a common earlier type." The Kendal Museum, of which Martindale was honorary curator, owes a great deal to his loving care of the herbarium, much of his valuable time having been spent in the preservation and arrangement of the fine collection there of the flowering and flowerless plants of the county. E. M. Holmes. SPAETINA TOWNSENDII Groves. [The fifth volume of the Proceedings of the Bournemouth Natural Science Society contains a paper by Dr. Stapf on the above-named plant, originally delivered by him before the Society as a lecture in 1913. Dr. Stapf's previous paper on this interesting grass was reprinted in this Journal for 1908, pp. 76-81 : the present contains much additional matter of interest as to the origin of the plant, some of which we here reproduce. The paper is illustrated by figures of S. alterniflora, S. stricta, and S. Toion- senclii. — Ed. Jouen. Bot.] Yabious theories have been advanced to explain the first appearance of the grass in the English Flora. The most plausible would seem to be that it was due to accidental introduction from a foreign country ; but our present knowledge of the genus and its distribution does not support it. Another suggestion is that Townsend's grass arose as a sport or mutation from SjMrtina stricta, which formerly used to grow on the shores of Southampton Water. Spartina stricta is, however, a singularly uniform and conservative species throughout its area, rather receding than advancing, and slow in adapting itself to changed conditions. It is evidently not the material from which one might expect sports or mutations to spring, so distinct and vigorous as Townsend's grass. There is, however, a third theory which is more plausible. According to it, Townsend's Spartina arose from a cross between >S'. alterniflora and S. stricta. S. stricta does not at present occur in the neighbourhood of Southampton or in Southampton Water ; but we know for certain that it did so not very long ago. S. alterniflora is common in the Itchen Eiver and also found in various places at the head and on both sides of Southampton Water. There was, no doubt, sufficient opportunity for the two species to hybridize. Unfortunately, it has not been possible so far to produce artificial hybrids of S. alterniflora and >S'. stricta. The evidence in favour of this theory is, therefore, necessarily circumstantial. It rests partly on the structure and the general behaviour of the grass, and partly on the occurrence of a natural 246 THE JOURNAL OP BOTANY hybrid between the same two parents in another part of the world and its extreme similarity to Townsend's grass. As to structural characters, there is no doubt that many of them may be considered as intermediate between those of S. alterniflora and S. stricta, although they are frequently, more or less, obscured by the remarkable readiness with which Townsend's Spartina I'esponds to external conditions, now dwarfing down to the modest size of S. stricta, now running up to and even exceeding the height of fine examples of S. alterniflora. Similarly, its remarkable vigour, its pronounced instability, and its varying fertility, very much enhanced in certain years and almost suppressed in others, may be adduced in favour of the hybrid nature of the grass, as those conditions are traits frequently observed in hybrids. But the strongest evidence seems to be in the following fact : — Spartina alterniflora and S. stricta meet out- side their English area only in one other place, namely, the estuary of the Bidassoa River, south of Bayonne, in the Bay of Biscay. There they grow intermixed, and among them has been found their hybrid. Foucaud described it in 1895, and named it Spartina Neyrautii, after its discoverer, Neyraut. Now this S. Neyrautii is so similar to S. Toivnsendii that Foucaud pro- claimed both as hybrids from the same parents, explaining such differences as there are by the assumption that S. alterniflora was the female parent in the case of the Bidassoa cross, and S. stricta in that of the English plant. The fact is very remarkable, and the argument deducible from it for the hybrid origin of Towns- end's grass has almost the force of experimental proof. Thanks to its vigour and occasional fertility, Townsend's grass has, in a comparatively short time, conquered thousands of acres of bare mud-land, it has invaded and, in places, much reduced the beds of Spartina alterniflora in Southampton Water, and even attacked the marshes which so far have been the home of Spartina stricta. However, its principal domain is and will probably for ever be the mudflats from one to three feet below high water-mark. Here the changes brought about by Towns- end's grass are remarkable. It is not only that the aspect of the flats is altered, the eye meeting great expanses of green com- parable to meadows or cornfields, where there was previously a monotonous sheet of grey at low- and half-tide, also the animal life on the flats and their physical character is undergoing a change. To mention only a few economically interesting effects on the fauna : in more than one place the larger molluscs which were collected for food have disappeared ; with the arrival of the grass, eel-spearing has been seriously interfered wath, whilst even duck shooting has been spoiled owing to the birds finding a welcome cover in the dense grass belts. But the most important change concerns the physical condition of the flats. It is obvious that the copious systems of roots and stolons must contribute to the stabilisation and solidification of the mud. In addition to this binding action the stems and lower leaves and leaf-bases act as a very effective strainer on the water, which is charged with SPARTINA TOWNSENDII 247 solid particles brought clown by the streams, catching and precipitating them. The result is an accelerated and increased deposition of mud over the area tenanted by the grass. The level of the mudbank becomes raised, the mud itself firmer. Further, the decay of each year's growth enriches gradually the mud with nitrates and sulphides and other salts, and prepares it for the reception of types of vegetation which were until then excluded from it. On the land side of the Spartina belt, where there is only a foot of water at high-tide, a growth of Aster Tripolium and Ohione portulacoides springs up among the grass, the first heralds of the reclamation of land that has set in. If the process continues, the muddy foreshore will gradually be replaced by terra firma. But another effect is more immediate, that of the protection which the grass affords to the shore behind it against the erosive action of the sea. The stems of the grass opposing themselves in their millions to the onrushing tides, to currents and the wind-driven sea, act like a natural breakwater to the shore behind them. It might be feared that the grass would become a nuisance to navigation by blocking up the waterways, but this is not the case. Bound to shallow water, it is not likely to invade the deeper water channels. On the contrary, the consolidation and gradual elevation of the grass-grown fiats along them tends to increase the scouring action of the currents and tides on the sides and bottoms of those waterways, making their banks steeper and increasing their depth. There is no reason why artificial plantations of Townsend's grass, under conditions corresponding to those of its native habitat, should not be successful. Propagation by division is easy, and the grass takes on well and grows rapidly, as experi- ments made in the Medway River and in New Zealand show. When the grass is young, the leaves and stems are succulent and sweetish, and cattle and horses relish it. Several American species of Spartina are cut and fed to horses and cattle on a large scale. Analyses of Townsend's grass, made on behalf of the Board of Agriculture, show that for nutritious qualities it is quite equal to its American allies, and may be classed as a good average fodder grass. Other uses to which the grass has been put and might be put on a larger scale are for thatching, and, above all, for mulching. It has even been tried for paper-making, but with doubtful success. SEX CHARACTER IN PLANTS. [The following interesting summary of the experiments at Merton Park appeared in the Tifues of July 20.] Work of great interest is now being done at the John Innes Horticultural Institution at Merton Park, where Professor Bate- son and his staff are conducting investigations in genetics and in the problems of sex characters and hybridization in plants. The whole question of variations and mutation and the transmission of sex characters from one generation to another is not only one 248 THK JOUKNAL OF BOTANY of the mosfc fascinating of the day, but it may, perhaps, have the largest importance to humanity. The work at Merton Park is not scientific, in the sense that it has no immediate apphcation to practical affairs. When we interpret it into an effort to produce a truly disease- resisting strain of potatoes, to grow flax a foot or two taller than it has been grown before, and investigate the farmer's curse of thrips, to increase the fertility of fruit trees, to turn out beautiful new varieties of well-known flowering plants, then the work seems practical enough. The work is young yet ; but every step gained, almost every series of experiments, adds some con- tribution, if only a negative one, to our economic knowledge. It is a pity that John Innes, who left his bequest for the foundation of a horticultural institution (and possibly had never heard the name of Mendel), cannot see to what excellent use his legacy is being put. If you go into the fruit house at Merton Park you will find it full of fruit trees — apple, plum, and cherry — from three to five feet high, growing in pots. Certain kinds of these trees have been known to be self- sterile — that is to say, that they cannot be fertilized with their own pollen but must be fertilized with that from other varieties. Also it has now been discovered that some distinct varieties are not capable of inter-fertilization. It is evidently of the first importance to fruit growers to know what varieties when crossed produce the best results. You will see here a tree, perhaps a cherry, which a month or two ago was a mass of blossom. There is a photograph to show what it looked like when every branch was covered equally densely with flowers. Now out of seven or eight branches five or six, it may be, are absolutely devoid of fruit. Two branches only are weighed down with clusters of ripe cherries. When the tree was in blossom the flowers on each several branch were carefully dusted with pollen from some other variety of cherry. The result shows which crosses were fertile and which were not. The method is not new ; it has been developed in the United States, but the results obtained here are and will be full of interest to British fruit-growers. In the flower houses sex investigations are being carried on by the crossings of begonias, calceolarias, nasturtiums, primulas, campanulas, and other flowering plants, and in calceolarias, especially, some quite new combinations of form and colour have been developed. Most of these are the results of experiments with C. cana, an unattractive, primitive-looking thing with woolly leaves which only a botanist would guess to be a calceolaria. So far as is known, G. cana has not heretofore been used in crossing ; but some of the hybrids from it are of great beauty, tall branching plants of the " tree" type, of novel shades of mauve and lavender and other curious tints. From the experiments with nasturtiums, again, some con- spicuously handsome double flowers have been produced ; one especially of a superb crimson-scarlet, and another almost equally handsome, banded with scarlet and yellow. In crossing the SEX CHARACTER IN PLANTS 249 single females with double males, the doubles produced are herma- phrodite males and sterile females. The female parent being single, fertile females carrying the "double" character apparently do not result, at least to the second generation (F. 2). There seems to be some likelihood here of a clue to the nature of the double 10-weeks stock, of which, also, the double flowers are presumably sterile females. Meanwhile, some of these new flowers, mere by-products on the line of investigation, are singularly desirable flowers. It is not possible here, even in outline, to indicate the trend of many of the experiments which are being conducted, the conclu- sions from most of which are at present most tentative. In working with begonias, after some years of experiment, the curious discovery was made, to the surprise of the discoverers themselves, that the ordinary exhibition type of double begonia is in many cases, if not in all, female. Some of the specimens carry ovules free on the petals. The discovery has necessitated the wiping out of the results of some four or five years of investiga- tions which had been carried on on the supposition, as is the current belief, that the flowers were male. Some very interesting suggestions are made, again, by the results of experiments with the varieties of the liliaceous plant CJdorojjJiytuvi, which has leaves variegated in longitudinal stripes of yellow and green. From forms of which the middle of the leaf is yellow and the edges green it is found that the seedlings are yellow. Where the colours are reversed — green in the middle and yellow at the edges — the seedlings are green. That is to say, that the seedlings appear to carry the characteristics of the stem and midrib and not of the outer portion of the leaf. If this is constant, it would seem to throw a ray of light on the question of what portion of the soma or body of an organism it is from which the germ derives its character ; a matter which may have an obvious bearing on inheritance in many things besides plants. An exhaustive series of experiments has been made in coloura- tion, using the common snapdragon or Antirrhmum, some of the results of which have been already embodied in scientific papers by the investigators. Briefly it has been found that all the combinations and variations of colouring in Antirrhinums are derived from four pigments — namely, ivory, yellow, red, and magenta. All of these have been isolated and obtained in a pure form, and the chemical identification of them is now possible. The first-named two pigments have been identified with apigenin and luteolin respectively. This chemical interpretation of Mendelian factors may obviously be a matter of far-reaching importance. We can hardly imagine the possibility of tracing through the germ the principle which makes for longness or shortness in pea plants (as in Mendel's famous experiments), or which produces the rose comb or single comb in fowls ; but we seem to be getting within reach of some- thing more tractable when we deal with a common colouring matter of known constitution. The equation begins to be not all unknown quantities. 250 THE JOURNAL OF BOTANY SHORT NOTES. Ptilota plumosa and Henry Goode. — I had the pleasure of the personal acquaintance of Henry Goode, who is said by Dr. J. Cosmo Melvill (p. 107) to have collected Ptilota i)lumosa at Falmouth. He was a most enthusiastic collector, and for several years used to bring his algas to me to name when he was in doubt. He lived at Plymouth as a centre, from about 1860-70, but went occasionally for a week or two to Falmouth and Penzance and other localities for seaweeds, and was lucky enough to find one or two pieces of Carpomitra at Penzance, and, if I remember rightly, also Croiiania and Gigartina pistillata. He corresponded with algologists all over the world, and when he died left in his will, concerning his herbarium, that I was first to take all the speci- mens that I cared for, and that Mr. F. W. Smith, of Falmouth, was to have the remainder. The foreign algae I selected from his collection formed the nucleus of my collection of foreign algae now in Mason's College at Birmingham. I may say that Mr. F. W. Smith, who resided at Falmouth, sent Goode many beautiful speci- mens— he mounted specimens in albums for sale privately, and also sold loose specimens to collectors ; the names were ascertained either from books or from correspondents, and both in his collec- tions and in Goode's I often found specimens wrongly named, localities were often added afterwards by Mr. Goode from memory, as he often forgot where the specimens came from, when not labelled at the back by the collectors ; Goode generally wrote the name in front of his specimens. I doubt, therefore, whether any reliance is to be placed upon the fact that Goode's specimen was labelled " Falmouth." I have visited Falmouth several times, but never saw Ptilota plumosa there : on the other hand, the Isle of Anglesea is quite a probable spot for it. I have gathered good typical specimens of Phyllopliora Brodim at Penmon in Anglesea, and under the Menai Bridge, as well as Phlceospora suharticulata and ClicBtopiteris plumosa, northern algae which I have never seen in Devon or Cornwall ; also Cordylecladia and other southern algae at Penmon, so that evidently at this point the northern algae find their southern limit on the West Coast, just as Dcles- seria angustissima finds its northern limit just below Scarborough, on the East Coast and southern algte extend to Anglesea. Ptilota plumosa was recorded from Holyhead and Port Dafarch some years ago by Mr. J. E. Griffith, of Bangor, in his Flora of Anglesea and Carnarvonshire (p. 237), as growing on the stem of Laminaria digitata. The only satisfactory statement concerning the locality of an alga is when it is found actually attached to a growing plant, or a rock, as many weeds are floated for a very considerable distance before decaying, and floating algae are always doubtful records from the spot where they are found. Even West Indian seeds are washed up in the Hebrides by the Gulf Stream. — E. M. Holmes. Gaultheria Shallon in Surrey. — A specimen of this has been sent me by a correspondent, who describes it as growing on sandy soil at a high elevation on Leith Hill, Surrey, " apparently quite wild." — H. J. Eiddelsdell. 251 REVIEWS. A Monograph of the Genus Sabicea. By Herbert Fqller Wernham, D.Sc, F.L.S. 8vo, cloth, pp. 82, with twelve plates and text-figures. Price 6s. London: British Museum. 1914. When volume iii. of the Flora of Tropical Africa appeared (1877), the genus Sabicea comprised nineteen species, a number gradually increased until, with the publication last year of the Cata- logue of the Talbots' Nigerian Plants, forty-four species had been described. It speaks well for the thoroughness of Dr. Wernham's research that he has detected no fewer than sixty-two additional species, thus making the total number known to-day a hundred and eight. This result is embodied in the excellent monograph with its twelve well-executed plates now lying before us. The nineteen pages of introduction are full of interesting reading, and, provided the author's views on the derivation of the various groups are regarded as suggestions merely as to what may have happened — and this is all that is claimed for them — it must be admitted that they are plausible and preferred with much ingenuity. The main grouping of the species is founded upon the inflorescence, the earliest form of which the author supposes to have been the open cyme, from which has been derived the condensed head, and finally the head surrounded by an involucre of bracts. The difficulty in this matter of descent is that we do not know whether unchecked advance from the simple to the more complex really has occurred in the history of any group of organisms. Thus, to take one case — a case with direct bearing, as it happens, upon the point in question. The head of Composites is allowed by all to be the highest expression of effectiveness in floral arrangement ; yet there are genera, undoubtedly derived from GompositcB of normal type, in which the head is reduced to two or three florets, or even a single floret in a scattered inflores- cence, thus harking back to a very primitive state of things. The truth is, we know little at present about the phylogeny of Angiosperms, and still less about that of their genera and species, and so far ontogeny cannot be said to have proved of much use in enlightening our ignorance. This, however, should not deter monographers from giving us their conclusions, for speculation can do no harm, provided its true nature be kept in view. Dr. Wernham's key has been carefully constructed, and the species should be easily recognisable from it. We notice, too, a very good point, one unfortunately not always present in recent monographs, viz. the citation of the herbaria where the various species may be found. We cannot, however, refrain from mention- ing that in some cases the full nomenclature has not been given. For instance, a Brazilian plant was considered in the Phanerogamic Botany of the Matto Grosso Expeclition to be conspecific with S. novo-granatensis K. Schum. Dr. Wernham finds this to be a mistake, and he describes the plant under a new name without reference to the erroneous identification. This is, however, scarcely a matter of primary importance, and we may hope 252 THE JOURNAL OP BOTANY that in fui'ther monographs of Uuhiacea which we are promised from the same pen full citations will appear. But we are not disposed to be any less warm on this account in congratulating the author on the capital piece of work he has turned out. S. M. The Standard Cyclopedia of Hortic^dture. By L. H. Bailey. Vol. i. A-B. 4to cloth, pp. 602. 700 figures in text. New York : Macmillan & Co. 1914. 25s. net. This work, which is to be completed in six volumes to be illus- trated with coloured and other plates, with four thousand engravings in the text, contains contributions from most of the leading horti- culturists and botanists of America, all of them experts in the particular subjects on which they write. We learn from the preface that the work " discusses the cultivation of fruits, flowers, and garden vegetables in the United States and Provinces." In style, it is similar to the Cyclopedia of American Horticulture published fourteen years ago, but the treatment in the earlier work was confined closely to " the trade "-^to those plants " sold in the United States and Canada"; in the present, "the trade" is interpreted more liberally, and includes the species offered and supplied to American customers by many European dealers. The horticultural possibilities of the various States are outlined, and biographies of eminent horticulturists are given. The first part contains a synopsis of the vegetable kingdom, based on Engler's system, with a key to the families and genera arranged according to the Genera Plantarum Bentham & Hooker. It seems illogical to employ the two systems used in the same work. The real reason for the key being on Bentham and Hooker's system is that it was prepared for the former Cyclo- pedia: an additional reason is that "the now system is better adapted for showing relationship or likeness, while the old system is well adapted for bringing out differences : most of those who use this part of the Cyclopedia will probably be in seai'ch for differences/' A list of English equivalents of the Latin names and a glossary of botanical and horticultural technical terms are appended. We read in the preface: "It is impossible now to know how many wrong determinations, inaccurate and in- sufficient descriptions, and faulty judgments, have been perpetu- ated from author to author through long series of years. . . . The best that can be done in very many cases is to accept the name appearing in a catalogue, and to attach to it the most authentic or most adaptable description of a recognised botanical species of the same name; there is no telling whether the dealers' plant is properly determined or whether it represents the botanical species bearing the same name." It would seem that the scientific horticulturist meets with the same difficulties in America as we do in this country. There is no reason why a nurseryman should not have pet names for his plants, even though these may have some semblance to real botanical names ; but the free manner THE BANANA 253 ill which names are printed in catalogues has resulted in what can only be described as chaos. Names are given which have no application to the plants described, and sold as such : the names of others sometimes look as if they were taken from one of the old herbals. A condensed description of each genus is given, followed by the native localities. The descriptive portion is characterised by the excellence which we always expect, and never in vain, to find in Mr. Bailey's work. Keys to the species are added, arranged primarily to aid the gardener in making determinations. The species are arranged systematically; and, where necessary, an alphabetical index is supplied for rapid reference : a list of synonyms and refer- ences is appended. The descriptions are excellent : although the editor hopes "that every entry will be worked over and improved within the next decade," the book is as authoritative and excellent as it is possible to be. The compilation is comprehensive, and the treatment of modern theory and practice exhaustive. The book is clearly printed on good paper, and well and neatly bound. The illustrations are clear and to the point, though some are of little artistic merit. As a Cyclopedia it stands far above anything we have seen, and the publication of the succeeding five volumes will be looked forward to with interest. J. K. Eamsbottom. The Banana : its Cultivation, Distribution, and Commercial Uses. By William Fawcett, B.Sc. 8vo. Pp. xi. 287, tt. 17. London: Duckworth. 1913. Price 7s. 6fL Mr. Fawcett is to be congratulated on the appearance of a useful and much-needed handbook. It embodies the experience gained during his twenty-one years' residence in Jamaica as Director of Public Gardens and Plantations, a period during which the value of the bananas exported from the island increased from £250,000 to £1,000,000. To the results of his own experience Mr. Fawcett has added those of an exhaustive study of the literature of his subject, including the cultivation of the fruit in the Tropics generally. The banana is the product of cultivated varieties of three species of Musa — M. sainentum, cultivated extensively in Jamaica and Central America and also in Malaya ; M. Cavendishii, the smaller and more delicate-tasting Canary Island banana ; and M. acuminata, grown in Malaya. The plantain, a much larger fruit, used cooked before it is ripe instead of bread or potatoes, is the product of another species, M. paradisiaca. Plantains are not exported to Great Britain, and, as Mr. Fawcett remarks, " they are not much wanted in countries where potatoes are plentiful and much cheaper, and are more valuable than bananas in the countries where they grow." After a preliminary chapter giving a full account of the structure of the plant-root, stem, leaf, tlower and fruit, a number of chapters are devoted to its cultivation, including the question of manures and fertilizers, with a short 254 THE JOURNAL OP BOTANY chapter on the financial aspect and some advice to those about to cultivate. Fungus diseases and insect-pests are described in detail and remedies suggested. The economic use of the banana provides material for several chapters, from which, apart from its well-known value as a food, it appears to have medicinal value, and also to be worth consideration as a source of alcohol. The author does not, however, encourage any hope of its competing with M. textilis and other species as a soui'ce of fibre. An inte- resting account is given of the development of the banana trade and the manner of transport of the fruit by sea and land ; and a useful series of chapters is devoted to a general review of the cultivation of the banana and plantain in various parts of the Tropics. The last chapter is a systematic botanical account of the species of Miisa, nearly seventy in number, each of which is briefly described. An appendix supplies a few recipes for cooking bananas. ABE Die Silsswasscr- flora Deutschlancls, Osterrcichs tmd der Schiveiz. Heft 1 : Flagellata I., von A. Pascher und E. Lammermann. Jena. 1914. This volume will prove a useful and reliable key to the Pantostomatinae, Protomastiginge and Distomatinse. It is a fitting complement to Part II. of the Flagellata already issued. Pascher contributes the introductory remarks and Lammermann the important systematic part. The figures are reasonably good, but the descriptions are in many cases very brief, with no critical remarks and no distribution. One fails to see why scores of species should be included in a flora of Germany, Austria and Switzerland, with the remarks " Bislang nur aus Nordamerika," &c. It implies that all Flagellates have a world-wide distribution, whereas there is much evidence to the contrary. Heft 6: Chlorophyceae III., von W. Heering. Jena. 1914. — This part includes the Ulotrichales, Microsporales and CEdo- goniales, and it is the best of the series yet published. The descriptions are good, but the figures, more especially in the Chgetophoraceae, leave much to be desired. The general account of the genus Microspora is incorrect, both in the cytology and formation of zoogonidia. Once more many genera and numerous species are included which should have no place in a Flora of Central Europe. Some American and even African genera and species are described and figured, none of which are known to occur, and most of which are never likely to occur in Europe. Inclusions of this kind are positively harmful and misleading, although apparently quite in keeping with the German character. Dr. Heering's treatment of the species of many of the genera is excellent and is accompanied by considerable critical observation. On the whole, this little volume will prove a useful, if small, laboratory guide. G. S. West. 255 BOOK-NOTES, NEWS, dc. One of the recently issued parts of the Records of the Botanical Survey of India (voh vii. No. 1) is devoted to the first portion of what is evidently a very thorough and exhaustive study of the Flora of Aden, by the Eev. Ethelbert Blatter, S.J., F.L.S., Pro- fessor of Botany at St. Xavier's College, Calcutta. Beginning with a history of the botanical exploration of Aden, with an account of what has been done by previous workers (with brief biographies), the physical aspects of the region — area and position, geology, topography, and conditions of plant-life are considered : an account of the vegetation follows, which includes its general aspects, with notes on its origin in the Indo-African desert and the North African steppe, and of the means of dissemination by wind and water and by the agency of animals and man. The systematic portion, with synopsis of orders and description of species, will appear later. This part contains one large and excellent map and five illustrations of the town and people. The other recent parts of the Records contain reports on the Mosses of the Abor Expedition (1911-12) and on Mosses collected by Mr. C. E. C. Fisher and others in South India and Ceylon, by Mr. H. N. Dixon ; this (vol. vi. No. 3) contains two excellent plates and descriptions of numerous novelties. In No. 4 of the same volume Messrs. G. H. Cave and Mr. W. Smith write on the East Himalayan species of Alangium, and the latter describes new species from the Calcutta Herbarium. We regret to find that we neglected to notice two papers reprinted from the Transactions of the Devonshire Association for the Advancevient of Science, Literature and Art — surely some shorter name might have been devised? — for 1913. In one of them Miss C. E. Larter gives a careful account of the distribution of Viola in Devonshire, in which she has had the help of Mrs. Gregory for the Noniimiwn and of Dr. Drabble for the Melanium section : to the latter V. mcduanensis Bor. is added, which has not previously been recorded for Britain. In the other paper Mr. Hiern edits the Fifth Eeport of the Botany Committee, which contains numerous additional species and localities for each of the eight districts into which the county has been divided. The first of the series of six volumes entitled The Oxford Survey of the British Envpire (Clarendon Press) is devoted to the British Islands and Mediterranean possessions ; to this Dr. Moss contributes a general account of the distribution of British plants and correlates Forbes's and Watson's work with that of recent Continental authors, as well as with that of Mr, Clement Eeid. The most recent addition to the " Bibliotheque Scientifique Internationale " is a treatise on La Vie et la Lumiere, by Dr. Raphael Dubois (Paris, Librairie F. Alcan ; price 6 francs). The first chapter is devoted to " les vegetaux lumineuses " as repre- 256 THE JOURNAL OP BOTANY sented by the genus Photobacterium, on which the author has made various experiments ; it is illustrated by a curious " photo- graphie du buste de Claude Bernard eclaire par la lumiere des Photobacteriees. ' ' Mr. p. Diemer, of Cairo, has published a pretty little book containing some thirty pictures, reproduced from photographs of Sovie Desert Floioers by Mrs. Grace M. Crowfoot, to which are added brief descriptions based on those in Muschler's Manual Flora of Egypt. There is an interesting introduction dealing with the characteristics of the Flora. The price of the book is not stated ; the copy before us is announced in violet ink on the title-page as "Gratuit pour la redaction," and the publisher adds to his generosity by sending a ready-written review, of which we do not propose to avail ourselves further than by agreeing that " with this book in hand a good many of the commoner of the flowers growing near Cairo can be readily identified." Dr. Staff contributes to the supplement to vol. 50 of the Botanische Jalirhilcher an important paper on "The Southern Ele- ment in the British Flora." We regret that a paper of such special interest to British botanists should appear in a German periodical with which few, if any of them, are likely to be acquainted. A note in the Selhorne Magazine for August, signed " K. M. Styan," records the finding, in August, 1913, of " a tiny patch " of Asperula nitida, " not very far from the summit of Ben Nevis." The identification has been verified at the National Herbarium, where a specimen of the plant has been placed. A. nitida is a native of Asia Minor, " now cultivated somewhat in English gardens." Although the writer of the note thinks the view that it was planted there is " too ridiculous to credit," we have little doubt that its occurrence is to be thus explained : it is, w^e believe, well known that attempts have been made to establish plants upon the Scottish mountains as elsewhere, and this seems to be one of the few occasions on which the attempt has proved successful. A SERIES of papers on " The Flora of the Brent Valley Bird Sanctuary," by Mr. J. C. Shenstone, appeared in the Selborne Magazine during 1913. Vegetationsbilder von Kilimandscharo, by Dr. Gertrud Tobler- Wolff and Dr. Fr. Tobler (Jena, Gustav Fischer, 1914), forms Parts 2 and 3 of the twelfth series of the "Vegetationsbilder" edited by Professors Karsten & Schenk. The letterpress accom- panying the plates explains in an admirable way the various features of the district visited by the authors. The plates them- selves are well executed and, although encumbered with the embarrassing excess of detail and want of individual definition common to similar photographs, give an excellent idea of the scenery depicted. Special mention may be made of plate 7 {Lco- notis mollis Benth), plates 9 and 10 (giant Lobelias), plate 12 (Heaths), and plate 17 (Helichrysums). This is altogether a valuable addition to this series of botanical photographs. Journ. Bot. Tab. 534. HYDRILLA VERTICILLATA Casp. 257 HYDRILLA VEBTICILLx\TA Gasp. IN ENGLAND. By Arthur Bennett, A.L.S. (Plate 534.) HvDRiLLA verticillata Caspavy in Bot. Zeit. xiv. 899 (185G). H. ovallfolia Rich, in Mem. Inst. Par. xii. (1811). Udora verticillata Gorski in Eicii. Skizze. v. Lithuan., itc, 127 (1830). U. lithuanica Besser in Flora 11. Beibl. 12 (1832). U. occidentalis Koch, Syn. eel. 1. 669 (1837). U. pomeranica Reichb. Icones f. 104 (1845). This interesting addition to the British Flora has been found by Mr. W. H. Pearsall in Lake Lancashire, v.-c. 69 b, at Estwaite Water, growing with Naias flexilis — itself an occurrence of note — Elodea canadensis, Potamogeton Sturrockii, P. imsillus and CalUtriche autumnalis. The " Water " where the plant occurs is rich in aquatics, and Mr. Pearsall suggests that this may be due, among other causes, to the number of water- fowl that frequent it — wild ducks, coots, water-hens, &c. — by the score. The English examples seem to be produced from winter buds, the first leaves being opposite, then gradually producing leaves in threes and fives (occasionally fours) at each node ; they are linear-acuminate, pellucid, with small cells on the leaf- margins which can hardly be called serrations, as they seem to be extra-marginal — not as shown in Elodea in English Botany, ed. 3, ix. t. 1446. The following description is given by Sir J. D. Hooker {P'lora of British India, v. 659) : — " A submerged leafy dioecious herb. Leaves short, 3-4-nately whorled, or the lower opposite. Male floiuers solitary, stoutly pedicelled, in a subglobose sessile muricate spathe ; sepals 3, ovate or obovate, green ; petals 3, oblong or cuneiform ; stamens 3, anthers large, reniform, opening elastically ; pistillode small. Female floivers 1-2, sessile in a tubular 2-toothed spathe ; perianth of the male, but leaflets narrower ; ovary produced beyond the spathe in a filiform beak, 1-celled ; styles 2-3, linear, undivided ; stigmas 3, fimbriate ; ovules anatropous. Fruit subulate, smooth or muricate ; seeds 2-3 oblong, testa shortly produced at each end. " Plant forming large masses. Leaves \-^ in., 4-8 in a whorl, with a short sheathing one at the base of each branch and a short pair above this ; linear or oblong-linear, serrulate or entire. Flowers ^-J in. long ; perianth segments very variable. Fruit smooth or sometimes muricate, or (in Ceylon) squarrose with filaments above the middle. The male flowers escape from the sheath when mature, and float to the top of the water." I have followed Ascherson and Graebner (Syn. Mitteleur. Fl. i. 399) in adopting Caspary's name for the plant. The genus is near Elodea and Anacharis, and perhaps all three should be included under one. The plant has been found in Europe in Journal of Botany. — Vol. 52. [October, 1914.] u 258 THE JOURNAL OP BOTANY Pomerauia, S.E. Prussia, Russia (in tlie governments of Wilna, Kurland, and Witebsk) and E. Asia, Australia, Mauritius, Central Africa, and Madagascar. Whether it is indigenous in Europe is doubtful: Nyman (Consp. Fl. Europ. Supp. ii. 285 (1890)) remarks: " Patria hujus plantie est India orient., ubi frequens dicitur." If not indigenous, it is not easy to suggest how it has become distributed, unless the seeds become attached to the feet of aquatic birds, many of which i^ange widely over the world. The suggestion that it is carried in their crops seems not admissible, as Danish ornithologists have shown that, as a rule, birds in migration travel with empty crops. The plant occurs also in Tropical Asia, Tropical Africa, the Mascarene Islands, and Australia. The occurrence of Naias in the same lake is also remarkable. We now have this species in Ireland, Scotland, and England; for its distribution in Ireland and Scotland see Trans. Bot. Soc. Edinb. xxiv. 16 (1909). In Europe it is rare, being recorded in Finland !, North Russia, Sweden in Scania ! and formerly in Upland, North Germany (Binow-See and Paarsteiner-See), Pome- rania and Lithuania. It is generally distributed in the United States and Canada. Hooker (Stud. Fl. ed. 3, p. 439 (1884) ) gives " Asia," but Dr. Rendle names no Asian stations in his revision of the genus in Trans. Linn. Soc. (Botany), vol. v. The accompanying plate is from a photograph kindly supplied by Mr. W. H. Pearsall. THE INDIFFERENT COILING OF ARUM SPATHES. By Miller Christy, F.L.S. It is now over thirty years since I first observed the fact that the spathes of the common Cuckoo-pint {Arum mamdatum) are coiled or rolled indifferently either way in different individuals — a fact which I have not seen noticed in botanical literature, except in a brief note by myself published in this Journal for 1883, p. 237. The same is the case with the flowers of the common greenhouse "Arum-lily " [BicJiardia africana or Calla (Bthiojnca). The peculiarity is common, probably, to all the Aroidae. Later, I observed that the leaves of A. maculatum, on their first appearance above ground, are also coiled or rolled indifferently either way, and that all the leaves on any one plant are always coiled or rolled the same way as the flower-spathes on that plant. This is, after all, merely what miglit have been expected ; for a spathe is no more than a modiiied leaf or bract. It may be asked, perhaps, how it is possible, without great difficulty, to substantiate the statement that all leaves and spathes on any one plant ai'e always coiled or rolled the same way, seeing that the leaves regularly appear at least two months, and often more, before the tips of the spathes are visible above ground and that they uncoil very soon after their appearance. In reality, however, substantiation is easy; for, when even the earliest leaves THE INDIFFERENT COILING OF ARUM SPATHES 259 first appear above ground (which is often as early as January or February), the spathe is ah'eacly fully formed, though, of course, very small and still entirely below ground. By pulling up the plant and splitting it open, one can easily ascertain, even at this early stage, which way its spathe is rolled. The facts noted in connection with the aestivation of the leaves and spathes of the Arum have long seemed to me curious and anomalous ; for Nature does not, as a rule, leave matters of this kind to chance (so to speak). In most similar cases, she follows some definite rule for each species or each genus, as the case may be. Thus, among climbing plants, a majority (as the Wliite Convolvulus and the Scarlet Eunner) revolve their shoots and tendrils in one definite direction, while others (as the Hop and Honeysuckle) revolve in the opposite direction.* Darwin observes f that, in almost all climbing plants, members of the same genus revolve their shoots and tendrils in the same direction. The same is true, I believe, of the aestivation of the flowers and leaves of all plants having flowers or leaves which are coiled or rolled in their earlier stages — that is to say, in each species or genus, the flowers or leaves are regularly coiled or rolled either one way or the other : not either way indifl^erently. Again, among mollusca having helically- or spirally-coiled shells, a large majority of species coil in one definite direction, though some species coil in the opposite direction. In most species of mollusca, however, one meets occasionally with abnormal individuals coiled in that direction which is opposite to the direction normal in the species — a peculiarity which is much commoner, for some unexplained reason, in some species and in some localities than in others. | The case of the spathes of A. maculatum is, however, totally different from any of the foregoing cases ; for, as stated already, the spathes of this species are coiled or rolled either way in- differently. Similar cases are, I believe, rare in Nature. In view of these facts, it occurred to me, several years ago, that it might be worth while to undertake investigations with a view to ascertaining whether plants of the Arum, having leaves and spathes coiled either in one direction or the other, occur in Nature in about equal numbers ; or whether plants having their leaves and spathes coiled in one direction or the other are in a majority. Accordingly, I began a series of investigations, which I con- tinued at intervals for five years. During country walks or in odd moments wherever I happened to be (often in shrubberies attached to my own garden), I gathered spathes of the Arum, afterwards counting and recording the number of each kind I had * See Darwin, Climbing Plants, 2nd ed., pp. 23-35 (1875), and Dr. B. Daydon Jackson, Glossary of Botanic Terms, 2nd ed., p. 3G7 (1905). The majority revolve in tlie direction which Darwin calls (oj). cit., p. 33) "against the sun." t Darwin, oj). cit., pp. 33-34. I See J. W. Taylor, Monograpli of Land and Freslncater Mollusca of British Isles, i., pp. 23-2-1, 108-109 (1900). u 2 260 THE JOURNAL OF BOTANY secured. In all cases, I gathered one spathe only from each plant. Further, to secure a more reliable result, I abstained usually from gathering more than one spathe from two adjoining groups of plants, whenever it seemed likely that both groups had grown from the seed of one parent plant. In this way I have gathered and counted either spathes or leaves or both from no fewer than 1228 plants — a number large enough, I think, to enable one to arrive at a fairly definite conclusion on the point at issue. The results of my counting are set forth in detail in a tabular statement which follows. Before examining those results, however, it is necessary to find terms capable of expressing, definitely and without ambiguity, the two directions in which the spathes and leaves of the Arum are rolled or coiled. For this purpose, I employ the terms " dextral " and " sinistral." These are, I hold (for reasons I hope to set forth elsewhere), the only terms capable of being used in such case without liability to confusion. Let the observer imagine himself to be standing upright, within the spathe (or, better still, within the central stem or axis), facing and looking out of the opening of the spathe. If, in that case, the wing of the spathe, on leaving the central stem or axis as it begins to coil, passes first to the observer's right hand, that spathe is what I call " dextral." If, on the contrary, it passes first to the observer's left hand, the spathe is what I call "sinistral." (It should be remembered that, if the observer imagines himself to be outside the spathe and facing its opening, the conditions are exactly reversed.) The annexed illustration shows a dextral and a sinis- tral spathe of A. viaculatum, as well as a transverse section through the " bulb " of each, the sections being viewed, of course, from above. In each case A represents the central stem or axis (within which the observer is supposed to be), and B the point he is supposed to face as he looks out of the opening of the spathe. Turning now to the results of my counting, as shown in the tabular statement annexed, it will be seen that, of 1228 Arum spathes, gathered at random, on thirty-two different occasions between 25th April, 1909 and 10th May, 1914, from hedgebanks, roadsides, ditches, and woods, at various places, chiefly in Essex and Gloucestershire, 645 proved to be sinistral and 583 to be dextral — a majority of 62 for the sinistral spatlies. It is clear, therefore, that, among the 1228 spathes examined, the sinistral spathes stood to the dextral spathes in the same relation that 100 stands to 90. In other words, the sinistral spathes out- numbered the dextral spathes by about 10 per cent. It is worth noting that we get much the same result if we take the results of the thirty-two observations separately. Thus, no fewer than nineteen (or about 60 per cent.) of these obser- vations yielded a majority of sinistral spathes ; only eleven (or about 34 per cent.) yielded a majority of dextral spathes ; while in two the numbers of each kind of spathe were equal. The numbers of observations made and of spathes counted are, THE INDIFFERENT COILING OP ARUM SPATHES 261 admittedly, not large. Nevertheless, the numbers appear to be sufficiently large to prove conclusively that, in Nature, sinistral spathes are really and persistently more numerous than dextral spathes. It is significant that this is demonstrated by the results •iV L%'> " 74 33 41 + 80 100 27 April 13 ,, Banks ct ditches. 47 30 17 + — 100 57 28 „ 28 ) J Broom Wood. 76 34 42 — + 81 100 29 „ 28 Banks ct ditches. 59 30 29 + — 100 97 30 May 2 Saffron Walden. Chalk-pit. 54 31 23 + — 100 74 31 „ 3 Chignal St. James. Banks ct ditches. 70 38 32 + — 100 84 32 „ 10 " " 26 10 16 — + 62 100 1228 645 583 19' 11 100 90 and two eqi lal 32 1 * Decimals ignored : the nearest whole number taken. t All from one cluster: grown, probably, from the seeds of a single flower. THREE NEW COMrOSIT.E FROM PERU 263 be of interest, too, to ascertain by experiment whether the rolling of the spathe constitutes what is known as a Mendelian character. Herein, however, I propose to do no more than place on record the facts observed. The number of similar cases which occur in Nature is, I believe, small. Darwin records '■' that the Woody Nightshade [Solanum Dulcamara), a plant with exceedingly feeble climbing powers, revolves its shoots and twines indifferently in either direction. He mentions! also one or two other more or less similar cases among chmbing plants. THEEE NEW COMPOSITE FEOM PEEU. By Albert Hanford Moore and Spencer le M. Moore. During a recent visit to South America Mr. H. O. Forbes made a small collection of plants in Peru, and upon his return presented them to the British Museum. The list includes several GompositcB, all known species except the three here described. The first is a Spilanthes ; of this, as it was not determinable from the clavis given by Mr. Albert Hanford Moore, of Washington, in his monograph of the genus (Proc. Am. Acad. Arts & Sci. xlii. 521, 569 (1907) ), a specimen was sent to that gentleman with a request, kindly complied with, to furnish a description, if its supposed novelty should indeed prove a fact. For the others I am responsible. — S. M. ^^ Spilanthes iolepis A. H. Moore, sp. nov. Caule erecto minute pubescente foliis ovatis ca. 2-3 cm. longis 1-1-5 cm. latis pihs albis instructis dentibus aut serrationibus induratis, apice acuto, basi rotundata vel subacuta, petiolis ca. 0-6-1 cm. longis ; capitulis subovoideis aut subcylindricis conspicue irregulariterque violaceo- punctatis, pedunculis 2-6-5 cm. longis; involucri squamis numerosis violaceis pubescentibus ; achseniis valde ciliatis. Hab. Peruvian coast to 7000 ft. A S. leucantha H. B. K. (Nov. Gen. & Sp. PI. iv. 210, t. 370 (1820) ) maxime differt achaeniis ciliatis ; a S. ocymifolia (Lam.) A. H. Moore {op. cit. 531) involucri squamis numerosis in locum 6-8, foHis plerumque majoribus ; et ab ambabus differt capituUs violaceo-punctatis et involucri squamis violaceis, foliorum dentibus induratis. The earlier illustrations of Spilanthes ocymifolia show a larger number of involucral scales, as characteristic of S. iolepis rather than of S. ocymifolia \ \ L'Heritier even gives the number as twelve in the accoiiipanying text. The involucre in the Illustr. Genres has a stiff diagrammatic appearance, and I have observed * Climbing Plants, 2nd ed., pp. 20, 34, and 43 (1875). t Op. cit., pp. 33-36. t Bidens ocijmifoUa Lam. — Poir. Illustr. Genres, iii. '244, t. GG8, f. 3 (18'23) : .Spilanth'us alhus LHer. Stirp. Nov. i. 7, t. 4 (1784). 264 THE JOURNAL OF BOTANY that crudities in old drawings, especially when these have been sketched from the living plants, which were afterward not avail- able to the author at time of writing, frequently find their way into the text. In other particulars, including especially the size and typical shape of the leaves, both Lamarck's and L'ileritier's descriptions and figures are sufficiently close to the widely distri- buted plant commonly identified with them. The insistence of both authors on the leaves being entire or subentire is important. The leaves of S. ocymifolia are rarely toothed, and then have broad not hardened teeth, while S. iolepis has little indurated teeth or serrations. S. leucantha is readily distinguished by the entirely glabrous achenes ; S. ocymifolia by the small number of involucral bracts (4-8), and leaves which are on the average much larger. S. iolepis differs from both in the violet colour of involucre and spots of the disk (note the names " leucantha " and " albus " above), and the indurated toothing of the leaves. It is also more pubescent, the leaves producing characteristic, short, somewhat stiff, greyish or whitish hairs. Wedelia Forbesii, sp. nov. Caule herbaceo gracili sparsim ramoso hispidulo, foliis parvis subsessilibus lanceolatis obtuse acutis basi obtusis trinervibus margine subevanide denticulatis membranaceis pag. utraque sed praesertim inf. hispidulis, pedun- culis solitariis folia longe excedentibus gracillimis ipso sub capitulo incrassatis hispidulis, involucri phyllis ext. late oblongis obtusis ima basi coriaceis aliter herbaceis phyllis int. ext. similibus nisi paullo minoribus omnibus extus hispidulis, ligulis circa 9 bene exsertis, receptaculi paleis late oblongis acutis prope apicem denticulatis dorso prominenter carinatis apice puberulis tenuibus decoloribus costa nigra percursis, achaeniis oblongo-Hnearibus appresse setosis, pappo cyathiformi ore breviter ciliolato-lacerato adjecta arista satis rigida duplo longiore. Hab. Valley between Pacasmayo and Eail-head, 7000 ft. Folia 3-4-5 cm. long., 1-1-5 cm. lat., minute glanduloso- punctata ; petioli 2 mm. long., hispiduli. Pedunculi 8-16 cm. long. Capitula pansa 1 cm. long., 2 cm. diam. Involucri phylla ext. 8x3 mm. Eeceptaculi palese 6-6-5 mm. long. Ligulse ovatse, apice bidentatae, binervosae, 8x5 mm. Achaenia adhuc maxime cruda 2-5 x '75 mm. ; radii compressa, 1 mm. lat. Pappus -4 mm. long. The small, almost entire leaves and long peduncles, together with the pappus, are the chief feature of the species. Trixis (j Aplochl^n^i;) hexantha, sp. nov. Eaimis fruticosis sat tenuibus sursum foliosis pubescentibus dein glabrescentibus, foliis parvis petiolatis oblongis vel oblongo-lanceolatis obtusis vel obtuse acutis basin versus gradatim extenuatis firme membranaceis supra puberulis subtus pubescentibus, capitulis aequaliHoris 6- flosculosis pro rata mediocribus in paniculas breves corymbosas oligocephalas ordinatis, involucri campanulati phyllis 5 uniseriatis lineari-oblongis obtusissimis nisi obtusis pubescentibus apice barbellatis adjectis bracteis paucis (solemniter 3-4) linearibus THE ROGUE WALLFLOWER 265 obtusis pubescentibus pbylla suboequantibus, receptaculi alveolis ore pubescentibus, corollis tiavis ex involucro plane eminentibus horum labio exteriori ovato-oblongo 3-clenticulato inteiioii alte bipai'tito, acbyeniis linearibus basi leviter angustatis apice breviter contractis papillosis, pappi setis copiosis 2-senatis scabriusculis dilute stramineis. Hab. Valley between Pacasmayo and Eail-head, 7000 ft. Folia exempl. unici nobis obvii summum 3x1 cm., pleraque equidem circa 20 x 5 mm., in sicco griseo-viridia ; costa media supra plana, subtus maxime prominens ; petioli 4-5 mm. long. Paniculae 2-5 x 2-5 cm. Pedunculi proprii saepius 2-4 mm. long., pubescentes. Bracteae 6-7 mm. long. Involucri pbylla 7*5 mm. long., 1'5 mm. lat., coriacea, margine membranacea, dilute griseo- brunnea. Corollarum tubus anguste infundibularis, 7 mm. long. ; labium exterius 4-4-5 x 2 mm. ; labium interius 3'5 x "4 mm. Achaenia 5-6 mm., pappus 8 mm. long. Distinguishable at once from T. paradoxa DC. by means of the small, narrow leaves and the short involucres. THE ROGUE WALLFLOWER. By F. J. Chittenden, F.L.S. [Reprinted by permission from the Journal of the Boyal Horticultural Society, xl. 83-87 (August, 1914). In addition to the papers cited, reference may be made to a note in this Journal for 1882 (p. 282) by Robert Holland, and to a paper by Duchartre — " sur une monstruosite de la fleur du vioHer {Cheiranthus Chein) " in Ann. Sci. Nat. (Bot.) ser. 5, xiii. 315-339, t. 1.— Ed. JOURN. BOT.] From time to time there appears among wallflowers a rogue form apparently without petals and looking at a cursory glance as though the flowers had failed to open. This rogue form is not confined to any one variety, but occurs in both yellows and reds. The form has been known for a long time and has even received a botanical name, for A. P. de Candolle ''■'- describes it under the name Cheiranthus Cheiri y gynantherus, with the following diagnosis : " Antheris nempe in carpella mutatis." It appears to arise suddenly from time to time, but, as the observations to be described below show, it may possibly be that certain apparently normal individuals among wallflowers are so constituted that their seed necessarily produces both normal and rogue form. That is, they may be hybrids in the Mendelian sense and, so to speak, carry the characters of both normal and rogue forms. On the other hand, we have no evidence to show that the rogues do not arise suddenly as seminal sports. The malformation existing in the rogues is a very peculiar one, in which both petals and stamens are involved. The petals are * DC. Prodromus, I. p. 135 (1S24). 266 THE JOURNAL OF BOTANY reduced to oblong coloured pieces about the length of the sepals or a little shorter. A remark in Masters' Vegetable Teratology ■'- seems to infer that at times the petals may be developed normally. He says, "In most of the flowers of this variety the petals are smaller and less perfectly developed than usual." Brongniart f makes a similar remark : " Dans ces Clieirantlms monstrueux, qui etaient tres nombreux dans les parterres du Museum en 1841, les sepales et les petales existent dans leur position habituelle, mais en general les petales ne prennent qu'un developpement imparfait, ce qui signale immediatement les plantes qui sont le siege cle cette monstruosite." We have not, however, met with any cases in which normal petals were developed. It is almost to be expected that sucli cases would occur, for it is scarcely credible that a single character-determinant should produce the remarkable and dissimilar abnormalities which occur in both petals and stamens. The most extraordinary change, however, is in the stamens which are converted into carpels. As Allman \ has pointed out, there is considerable variation in the number of the supplementary carpels and in their adhesion. The full number is six, derived from the six stamens, but those corresponding with the two lateral stamens are not infrequently smaller than the others, or altogether absent. Allman found the ovary with the short style of these supplementary carpels was derived from the filament of the stamen, while " the stigma was plainly a transformed anther." Brongniart § has well described the various forms of this rogue met with, all of which we have seen in our own cultivations (see figure). We cannot do better than quote Dr. Masters' translation of his notes. II " Sometimes these six carpellary leaves are perfectly free, and in this case they spread open, presenting two rows of ovules along their inner edges, or these edges may be soldered together, forming a kind of follicle like that of the Colum- bine ; at other times, these staminal pistils are fused into two lateral bundles of three in each bundle, or into a single cylinder which encircles the true pistil. In a third set of cases these outer carpels are only four in number, two lateral and two antero- posterior, all fused in such a manner as to form around the normal pistil a prism-shaped sheath, with four sides presenting four parietal placentae, corresponding to the lines of junction of the staminal carpels." The conversion of stamens into carpels is a comparatively rare phenomenon, though conversion of stamens into petals is frequent. It occurs in Papavcr souinifcnon, and we have seen it in P. oricntale in our own garden, but in these cases, only some of the stamens are transformed ; it has also occurred in Polemonium * Masters, M. T., Vegetable Teratology, p. 305 (1869). f Brongniart, A., " Sur quelques cas de transformation des etamines en carpelles." Ball, de la Soc. Botaniqtie de. France, t. 8, p. 453 (1801). I Allman, Prof. G. J., " On the Morphology of the Fruit in the Cruciferae, as illustrated by a monstrosity in the Wallflower." Report Brit. AiSociation, July, 1851 (Ipswich) Trans., p. 70 (18.V2). § Brongniart. A.. I.e. ante. ]| Masters, M. T.. I.e. ante. THE ROGUE WALLFLOWER 267 ccerukum. Masters '■■ quotes Goeppei't as saying that the peculiarity in P. somniferum was reproduced by seed for two years in succession, but whether the seed was produced by the central or the supplementary carpels he does not say, while Brongniart \ obtained fertile seed of P. coarulenm from both central and supple- The " Rogue " Wallflower, with details of variation in the Flower Structure. mentary carpels. He does not record the result of sowing this seed, however. No one seems to have tried to obtain seed from the lateral * Masters, M. T., I.e. ante. t I.e. ante. 268 THE JOURNAL OF BOTANY carpels of the rogue wallflower until Professor G. Henslow, in 1910 or 1911, pollinated flowers of a rogue which occurred in his garden at Leamington. '■ Some of the flowers he pollinated from a red, others from a yellow variety. Both central and supple- mentary carpels set seed, the former much more than the latter. This seed was sown at Wisley and grown on to flower, the plants produced being all alike, except in colour, and all normal. None showed variation in number or form of petals, stamens, or carpels, but both red and yellow flowers were produced, some of the former with streaks of yellow. The normal type was thus clearly completely dominant to the rogue. We do not know to which colour type the original rogue belonged. Seed was saved from these plants interpollinated and sown as soon as ripe. Some of the resulting plants flowered in 1913 and showed that the seed had given rise to two types, the normal and the rogue, but as many had not arrived at flowering size they were all grown on to flower in 1914. A few plants died from one cause or another, but 143 flowered, and of these 101 were of the normal type (both red and yellow) and 42 of the rogue type (both red and yellow). On the assumption that we have to deal with a simple 3 to 1 Mendelian segregation, the expectation would be 107 normals and 36 rogues, and the numbers obtained are sufficiently near to the expectation to suggest that simple segregation is taking place. The case is a particularly interesting one, for the differences between the two forms are marked and complex, and the fact that the dominance is complete is in itself very interesting. As we have said, the change from maleness to femaleness is a rare one, but the results of the experiment seem to show that in CheirantMis femaleness is recessive to maleness. The persistence of this rogue type in small numbers, even though now great care be exercised in eliminating them from plants growing for seed, may be readily understood, if we assume the rogue on its first occurrence produced seed.f The seeds produced by it must have been hybrids, since the rogue itself produces no pollen, but they would doubtless have been sown among others from perfectly normal plants, and the culture would consist of many true normals, and a few hybrids, apparently normal, and quite indistinguishable from the normals in structure. The normals would far outnumber the hybrids, and the chances of interpoUination among the latter would be correspondingly small, with the result that, while hybrids would be produced with each succeedmg generation, rogues — the pure recessives — would rarely appear. We may show this graphically by the following diagrams, where N stands for the dominant normal, r for the recessive rogue. Crossing the rogue with the normal (which can be done only one way) we have : * Journal R.H.S. xxxviii. p. xxxix. (1912). t Roguea allowed to grow among plants produce a few seeds without artificial pollination. THE ROGUE WALLFLOWER 269 N (? X r ? Ni- j, (hybrid) F, Seed may be produced by the Nr plants either by intercrossing among themselves or by crossing from the normals, and the results will be different in these two cases : 1. Where the hybrids intercross : Nr X Nr NNf Nr| Nrf rr $ (normal) (hybrid) (rogue) and rogues will reappear in the proportion of one to three, two of which are hybrids. 2. Where the hybrids cross with the normal (the most likely thing to happen) : NN dagineum, Panicum repens, Andropogon dis- tachyum. Willkomm also credits him with {Juniperus Oxycedrus), Serapias cordigera, (S. Lingua), [Orchis longicruris), and Ophrys Vlll A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD aranifera ; but these are not in Schott's list, and may have been mis-cited for the Kock instead of from the neighbourhood. P. Salzmann collected in Spain in 1823. A considerable number of citations from his collections exists, many of them as " Gibraltar," but only four out of the following ftve not pre- viously recorded are established for the Eock, Sj^ergularia fimhriata having been more probably found on the Neutral Ground : — {Spergularia fivihriata), Ononis diffusa, Orohanche sanguinea, Piptatheriwi ccendescens, Scleropoa loliacca. Philip Barker Webb visited Gibraltar in 1827, and pub- lished about a dozen names in his Iter Hisimnicxtm in 1838. A few other species are also referred to him, but only the following are new : — Frankenia Icsvis, Silene moUissima, [PterosjMrtmn lasian- tJmm] , Lathyrus Clymenum, Vicia gracilis, Cyperus badius. Eambur worked at about the same time as Webb, but has few new records. Those which have been credited to the Eock are as follows : — Oxalis ceuiiia, Sarothamus bceticiis, {S. grandiflorus). Dr. K. FiNDLAY collected in the neighbourhood about 1835. I have seen the following specimens of his labelled Gibraltar, but it is not stated that they were found on the Eock : — {Lathyrus angulatusl), {Centaurea i^olyacanthal), Anagallis linifoUal (already recorded), {Armeria macrophylla !). E. BoissiER, in Voy. Bot. dans le Midi de I'Esp. (1837), gives seventy-seven species of his own collecting on the Eock. The following, which include a few cited elsewhere, have not been previously recorded. Those excluded belong to the Neutral Ground : — Fumaria parviflora, Malcolmia littorea, Succowia halearica, Brassica papillaris, Silene colorata, S. gibraltarica, Lavatera trimestris, Cytisus linifolius, Onojiis reclinata, 0. gibraltarica, Caucalis leptophylla, Daucns gijigidium { = D. gummifcr), [Cachrys PterocJdana] , [Pycnocomon rutcefolium] , Centaurea spJicBrocephala, Hedypnois arenaria, Thrincia hispida, Jasione montana, Solanum nigrum, Linaria pedunculata, L. amethystea, Nepeta tuberosa, Teucrium Poli^im, Achyranthes argentea, Bumex thyrsoides, B. tingitanus, Buppia maritima ! (as B. rostellata), Buscus Hypo- phyllum, Gastridium lendigerum, Ammophila arundinacea, Tri- setum neglectum. The following species are attributed to Boissier or to Boissier and Eeuter by Willkomm and Lange and others, doubtless from herbarium specimens, as they are not recorded from the Eock by those authors. I have seen Boissier's specimen of HelicJirysum decumbens. The stations cited on Boissier and Eeuter's labels are in most general terms and often misleading : — Fmnaria macrosepala, Limmi setaccum, Chcetonychia cymosa, Helichrysujn decumbens ! Picridium vulgare, Andryala integrifolia, A, arenaria, Adiantu7n Capillus-Veneris. A FLORA OF G.]5nALTAR AND THE NRIGHBOURHOOD ix M. WiLLKOMM p iblished in Bot. Zeit. 1845 an account of his travels in that yeca- through the south of Spain, whicli was translated in Ann. Nat. Hist. xvii. pp. 115, &c. He gives (on pp. 737-743 and 753, /. c.) the following new records for the Eock : — lianiincnlus flahcllatus, Geranium rotund ifulium, G. Boherti- anum, ErocUum moschatum, E. laciniatum, Ituta hracteosa, Pistachia Lentiscus, Physanthi/llis tetra'phylla, Psoralca hitumi- nosa, Hippocrepis multisiliquosa, Echalion Elatcrhan, Asieriscus spinosus, Pinardia coronaria, Pliagnalon saxatile, Senecio minutus, Galactitcs tomentosa, Centaurea pullata, Plcridium tingitanuvi, JEtlieorrhi.-ja bulbosa, Convolvulus siculus, Solanum villosum, S. Sodomceum, [CludnarrJnuum origanifolium] , {? C. villosum), Vero- nica Cymbalaria, Emex spinosus, Osyris lanceolata, Mercurialis annua, E^bpJiorbia helioscopia, Urtica membranacea, Oplirys lutea, Ornithogalum umbellatum, Smilax mauretanica. Willkomm gives a few further records in Prodr. Fl. Hisp., which was not published till 1870, and others are attributed to him by later writers, and though there is no doubt these were collected in 1845, I am not sure that they all came from the Rock. His earlier labels often leave it doubtful whether the specimens were gathered on the Rock or not. His records in the Prodromns would be more numerous but for the fact that he passes by many with the remark that they are common in South Spain : — {Malcohnia lacera), Fumana glutinosa, [Erodium Botrys, the specimen is E. laciniatum !] , Betania monosperma, Vicia hybrida, Scandix Pecten- Veneris, Vaillantia hispida, Senecio foUosus, S. lividus, Sonchus tenerrimus, Andryala arenaria, Echiwn calycinum, Stachys hirta, Plantago Psyllium. Dr. C. M. Lemann collected at Gibraltar in 1840-1. Most of his records were communicated to Kelaart, but two or three are cited by others, and I have seen specimens at Kew of those indicated. Those not previously recorded are : — lianunculus bUpharicarpus, Hesperis laciniata, Biscutella Icsvi- gata, Alsine temiifoUa, Spergida arvensis, Linum angustifolium, Lavatera cretica, Bhamnus Alaternus, Trifolium subterraneum, [Astragalus depressus'] , Ornithopus compressus, Lathyrus setifolius ! Thapsia villosa, Ferula communis (prolDably F. tingitana), Valeri- anella discoidea, Hyoseris scabra, Echium Pomponium, {Linaria viscosal), [Orobanche reticulata], Micromeria grceca, Thesium humile. Euphorbia serrata !, EpJiedra frag His, Orchis cordata, Fritillaria lusitanica, Uropetalum serotinum, Muscari comosum, Equisetum ramosum, Asplenium Adiantum-nigrum. Dr. E. F. Kelaart, in 1843-5, enumerated 174 native or naturalized species not previously noted. The list is too long to reproduce. The number does not include any doubtful records or cultivated species. H. A. Hurst collected in the neighbourhood in 1848. He was careful to label plants not collected on the Rock " near Gibraltar," Journal op Botany, Octorrr, 1914. [Supplement] p, X A FLORA OP GIBRALTAR AND THE NEIOHI50URHOOD SO I think the following, labelled " Gibraltar," may be relied upon : — Delphinium Staphisagrial, Atractijlis ccnicellatal, and Chlora perfoliata ! John Ball collected in the neighbourhood in 1851, and again in 1871. I can only trace the following, probably for the Kock: — Sisymbrium Trio and Antirrhinum tortuosum. E. Dautez collected from 1872 to 1880, chiefly in the San Roquo district, but lie has the following records from the Rock. His collections were all determined by M. Debeaux : — Fumaria agraria, Diplotaxis erucoides, Baphanus Puqjhani- strum, Lepidiuiii Draha, Senebiera Coronopus, Tuheraria incon- spicua, Dianthus Caryophyllus, Saponaria officinalis, Sagina maritima, [Erodium Sahmanni), Melilotus parviflcra, Trifolium suffocatum, Astragalus hamosus, Coronilla glauca, Lathyrus Ochrus, (Enothera stricta, Fedia Langei, Conyza ambigua, Gnaphalium luteo-album, {Senecio Icucanthemifolius), Hedypnois tubceformis, Pi- cridium intermedium, Sonchus asper, {Andryala laxiflora), Cusmita Epithynmm, [Lithospermum fruticosum). Salvia triloba, Lamium amplexicaule, Stachys lusitanica, Amaranthus deflexus, Rumex crispus, Parietaria mauretanica, Celtis australis, Ophrys Speculum, Allium iMniculatum, Setaria verticillata, Cynodon Dactylon, Scleropoa rigida, Lamarckia aurea, {Bromus tectorum), Ilordeum maritimum. G. Maw collected Crocus Sahmanni on Windmill Hill in 1883. M. A. DE CoiNCY gathered Lemna gibba in or near the Inun- dation in 1887. He also described in Journ. de Bot. xiii. p. 162 (1899) Trifolium carteiense from near San Roque. I have not seen specimens, but it seems to be near T. lappaceum. E. Reverchon collected on the Rock in 1887, but though his new records for Algeciras are considerable, he found nothing new on the Rock. A few records each have been assigned to the following, with their approximate dates, but the usual doubt exists as to the exact habitats, and I have only admitted them with doubt: — Funk (1848), Laguna (1860), Amo (1860-61), Fritze and Winkler (1873), Bilimek (1877), Hegelmaier (1878), and Perez Lara (1876-80). The present Flora shows about 101 new records of species, that is, those for which no definite former Rock records exist ; but some of them, from their commonness, must have been known to Kelaart and Dautez, and their omission was probably due to an oversight, and about ten were already known on the Neutral Ground. There are in addition about half a dozen naturalized aliens, or casuals, recorded for the first time. It would occupy too much space to give details of the records of plants found in the neighbourhood of Gibraltar. The botan- ists who have worked the region are practically the same as those who have explored the Rock, with a few additions. Very few of the records of Von Martins and Schott have been A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD XI taken up by later writers on the botany of Spain, while many of those of Salzmann, Broussonet, and later collectors have been adopted. The justification for this is not apparent, unless it be that modern writers liave included species from collectors which they have seen. I hesitate on my own responsibility to accept them all, especially as many of them are highly improbable, tlius their whole lists are open to suspicion ; but I liave included the more probable ones in the list of species to be searched for which is given at the end of this preface. Clusius worked a great deal in Spain, but his localities are not precisely defined, so there are no additional records to those on the Rock. Tournefort also published no list for the surrounding country. Schousboe (1798-9) first found Linaria Munbyana, Salvia bullata, and Narcissus viridiflorus, while Gaudichaud, who gives such an interesting list of Rock plants, did not visit the country round. Lists of plants collected by the following exist : — A. MS. list of Masson's plants in the National Herbarium names seventeen species for San Roque, of which Satureia Juliana is unconfirmed by later authors. Von Martius's list contains 192 species, at least twenty-five of which are improbable or wanting in confirmation. This large number may be due to incorrect diagnoses, or to the somewhat loose application of the older names, the synonymy of which is often doubtful. Others which are confirmatory of records I had excluded, as well as some of the more likely ones, are taken up in the Appendix to this work. Schott's list, which he heads "plants collected," contains names of species only known for the Rock, as well as some only known for the neighbourhood, so there is some doubt as to whicii they should be referred to. It contains 150 species, with about the same number of doubtful records as that of Von Martins, to which the same remarks apply. Salzmann only names five species, of which only Ononis hispida, for which he may have mistaken 0. Cossoniana, is unrecorded elsewhere. Willkomm and Boissier published lists of some of the plants they saw in their travels in the works already referred to. These are all taken up in the present Flora. Kelaart's list in Flora Calpensis contains, as already explained, species which occur beyond our limits ; those which are uncon- firmed for our region are shown in the Appendix to this work. Dasoi, who collected in 1887, submitted his plants to Gandoger for determination. The result is the inclusion of many im- probable names. Moreover, there is no knowledge of the extent of country explored by him, so that his list has no value. It was published in Bull. Soc. Bot. Fr. xxxiv. pp. 223 and 309, and some account of it will be found at the end of Debeaux's Flora. A much more trustworthy list of Reverchon's finds, mostly about Algeciras in 1887, is published by Rouy in the same volume (p. 434). Fritze, Winkler, Nilsson and Hackel also found B 2 Xll A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD many interesting species at about the same time, but I have seen no separate publication of their discoveries. Dautez, between 1872 and 1880, did an immense amount of good work chiefly in the San Eoque district. His plants, many of which have been found by no other botanist, were detei'mined by Debeaux. Of other collectors not previously named, mention may be made of Gutierrez, La Gasea, Alioth, de Noe, Kusisnky, and Porta and Rigo, the two last-named finding some half-dozen North African species not hitherto found in Europe. Literature. Three works have been devoted to the botany of Gibraltar and its neighbourhood, and one to that of the whole province of Cadiz. The Flora Calpensis, by Dr. E. G. Kelaart, M.D. (1818 ?-1860), of the Army Medical Staff", published in 1816, was for many years the only book on the subject. He enumerates 512 species for the Eock ; this number includes about a dozen species mentioned in the notes in parts II. and III., which have been confirmed by other authorities. Several other species so mentioned have not been included. He also enumerates 229 additional species for the neighbourhood, i. e. within twenty miles of Gibraltar. His Gibraltar list must, however, be largely discounted by the exclu- sion of many cultivated species, obvious errors or at least very doubtful records, and naturalized aliens, as well as of plants recorded as species but now reduced to varieties. His nomencla- ture is sometimes difficult to follow, and there is usually much doubt as to which species were found by him on the North Front, and which on the Neutral Ground, which I have treated as a separate district, and which is really a part of Spain rather than of Gibraltar. Kelaart uses the name Neutral Ground to cover both areas, and though, doubtless, many of the species now found only on the real Neutral Ground were in his day also found on the North Front, there is no certainty ; hence some of my exclusions from the restricted Flora of Gibraltar or Rock (District I.) may appear arbitrary. The exclusions from Kelaart's list under these various headings are : — Cultivated or casual species, and those now reduced to varieties ..... 53 Naturalized aliens ..... 9 Probable errors (some excluded, some only marked with doubt in this Flora) . . 42 Confined to Neutral Ground . . . .12 Leaving 396 species native on the Rock, having the same status as those admitted into the present Floi'a. In my citations from Kelaart's book I have made use of the details given in Part II. of tlie work, which amplify or modify those given in Part III. After an interval of forty years, Don J. M. Perez Lara began in 1886 the publication of his Florida Gaditana — the whole A FLORA OF GIISRALTAR AND THE NEIGHBOURHOOD xiii pi'ovince of Ciidiz — in Anales Soc. Hist. Nat. Espan. His work runs interruptedly through vols. xv. — xxvii. (1898), and was com- pleted in 1903 by an appendix published in the Mem. Soc. Esp. Hist. Nat., vol. ii. Ho does not include many of the previous records, but the later ones of Eeverchon, Winkler, Dautez, and other collectors, are all noted. The Florida contains useful notes on some of the species, though their utility would have been enhanced had more attention been given to the difi'erentiation of allied species and varieties than to a mere statement of their range of variation, which is confusing rather than elucidatory. Gibraltar being geographicallly considered as part of Spain, no special effort has been made to segregate its records, conse- quently several for the neighbouring country have been credited to the Eock itself, and sometimes the reverse mistake has been made. This is doubtless due to the comprehensive use of the name "Gibraltar" by most collectors, and the ambiguity of the name " Neutral Ground," though not actually used by Perez Lara, also has led to confusion. Perez Lara's work was a great advance on Kelaart's, and surpasses in utility and number of local records that of Debeaux, though the latter was specially devoted to the flora of Gibraltar and its neighbourhood. The total records for the whole province are 1905 species, and after making due allowance for excluded species, doubtful records, naturalized aliens, and those reduced to varieties, there are the following records for our region : — For the whole region . . 951 For Gibraltar only . . 462 These totals are doubtless larger than they should be, because I have included all species known for our region which Perez Lara specifies as common in the whole province, without giving any detailed stations. Some of these are certainly rare with us, a few are not recorded, and quite a number are not known for the Eock ; these last I have, of course, not included in the totals. It is highly probable that no station within our limits was known to Perez Lara for many of the others, but I have been unable to discriminate, so had no alternative but to include all that I knew to exist with us. A Synopsis de la Flore de Gibraltar was published by 0. Debeaux in 1888 in Actes Soc. Lin. Bordeaux, vol. xlvii. The author was but slightly acquainted with the region, and appears to have done little or no collecting there himself, relying mainly on notes and specimens submitted to him by Dautez : consequently he gives quite a false impression of the relative frequency of the species, as well as sometimes wrongly assuming a variety only to exist, whereas the type is far the commoner. His prefatory remarks also lose much of their value from the obscurity of his topography ; tluis, he confuses St. George's Hall, on tlie northern precipice, with St. George's Tower, near the southern end, better known as O'Hara's Tower but now demolished. I have not, therefore, availed myself of the special localities given in that XIV A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD part of his book. In some cases, also, localities are incorrectly cited, but usually this is due to the comprehensive nature of those on collectors' labels referred to above. There are many omissions of localities recorded by Kelaart for the Eock ; though he cites the page in Kelaart's Flora in which the records occur. On the other hand, he credits that author with many localities not mentioned in his work, apparently assuming that Dautez's and Kelaart's stations were identical. Many of my records for District I. have been due to acceptance of Debeaux's citations which further research has proved to be false, necessitating many corrections in my Appendix. Reverchon's finds are often only mentioned in observations interspersed in the text, and it is not clear why these and others are not included in the numbered series. Debeaux enumerates 965 species, and an additional 72 in his observations, making 1037 for the whole region, of which only 367 accepted species are mentioned for the Rock, 29 less than Kelaart enumerates, after making similar deductions for exclusions, &c. In 1910, Mr. B. H. T. Frere published his Guide to the Flora of Gibraltar, which is almost entirely a compilation of previously published records, with the descriptions of all non-British species added. Unfortunately the author had overlooked the publication of Perez Lara's Florida, so a large number of available records are omitted, and the few that are added are mostly for species which occur far beyond our limits. The descriptions are not very clearly drawn up, since in attempting to make the work popular the author has made a compromise between technical and popular language, acceptable neither to botanists nor to those with no botanical knowledge. The omission of the descriptions of British species necessitates constant reference to other works which are not always readily olitainable on the spot. There is no attempt to emphasize the differences between allied species, and scarcely any varieties are given. The stations are very vaguely cited, no personal verifications are given, and there are no authorities for the nomenclature. As the work contains very little that is new, and omits Juncacece and all later orders, I have not attempted to tabulate the species recorded. The Botanical Districts. I have divided the region into three main districts : — I. Gib- raltar, i. c. the Rock and the North Front as far as the British Lines. II. The Neutral Ground, kept as a distinct district for reasons explained below. III. Spain, which I have subdivided into — i. the San Roque subdistrict, from the Mediterranean to the Guadarranque River, as far north as the limits of the region ; ii. the Algeciras subdistrict, or all that portion of the region lying south of the Palmones River ; and iii. the Palmones subdistrict, between the rivers. District I., Gibraltar, is of special interest, being wholly British territory. It consists of the Rock proper, with that part of the isthmus joining it to the mainland called the North Front. A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD XV The latter is sandy, and is now levelled and turfed for recreation where it is not occupied by buildings ; only a small portion near the Devil's Tower and the rifie butts is in anything like its original state. Formerly the North Front was for the most part as wild and rough as the present Neutral Ground, and produced many species now " improved " out of existence, thougli some doubtless survived until Kelaart's time. These will be dealt with more fully under District II. The Eock itself is a mass of limestone, sloping towards the west, with the bare rock appearing all over the middle and upper slopes. The backbone of the Eock, which is about two miles long, rises vertically from the North Front to an altitude of 1350 ft. into a sharp ridge from 1000 ft. to 1440 ft. high, extending to O'Hara's Tower, whence it falls steeply to the plateau of Windmill Hill, some 250 ft. above the sea and thence precipitously to Europa Flats at about 100 ft. The shore, except at the North Front and Catalan Bay, and a few points on the western side, is precipitous or artificial, consequently but few maritime plants are found, these being chiefly at Europa Point and Catalan Bay. The slopes are covered with much shrubby vegetation, often difficult to penetrate, and in recent years a considerable number of pine and other trees have been planted on the w^estern slopes. The Alameda Gardens and the slopes towards Eosia were formerly much wilder, and parts of them were cultivated for vines and cereals. In this neighbourhood doubtless several of the species which are now extinct were to be found. On the east side a steep slope of sand gives a foothold for several species which only occur there and on the Neutral Ground, but doubtless formerly also grew on the North Front. Modern improvements and water catchments are rapidly destroying this interesting habitat. Bras- sica painllaris has already disappeared from its only known station. Gibraltar, which has 587 native species, has a markedly diffe- rent flora from that of the neighbouring parts of Spain. This is due to its limestone formation, the absence of water or damp places, the rarity of sand dunes, and the almost total absence of agriculture. Thus Ericacea are totally absent, while many other genera and species, which are quite common in the neighbour- hood, are absent, or at least very rare, on the Eock. There are at least 236 such species, without reckoning moisture-loving ones. Another cause, which, however, has tended to a reduction of the number of species ratlier than to their nature, is to be found in the alteration to tlie western slopes. Formerly these slopes were open to the browsing of goats, so that the shrubs were rarely allowed to rise to a greater height than three or four feet, allowing plenty of room between them for smaller plants, but now the planting of trees and the erection of an unclimable fence at a low elevation, have kept out the goats, so that the taller growths have been encouraged and crowded out much of the undergrowth. Gaudichaud remarks upon the total absence, in 1817, of all trees taller than Chanicerops luDnilia, though this must have Ijocn an XVI A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD exaggerated statement. Europa Flats and the south and south- east slopes have been less affected by this cause than those to the west and north-west. On the other hand, forty-eight species found on the Eock are absent from the adjacent territory, the following not being recorded from the Cadiz province, some of them (shown in roman type) not being known in Spain at all. No doubtful records or naturalized aliens are included in this and similar lists : — Clematis cirrliosa v. Dautezi (endemic), Hcsperis laciniata ? Brassica sabularia v. papillaris (endemic), Iberis gil^raltarica, IlcUanthemuuh origanifolium, Diaiithus Caryoijhijllus, Silene gibraltarica, Spergularia riipcstris, Linam niaritimuin, Lavatera arborea, Mediccujo ol)Scura,]\£eUlotus iiifesta, TrtfoUtnii suffocaliivt, Seduni viicrccnthum? Saxii'raga globulifera v. gibraltarica, Ferula tingitana? Vaillantia muralis, Hclichrysum rupestre v. Boissieri (endemic), Senecio miuutus v. S'. commutata Guss. Similar places ; rare ? 3-5. Like last, but leaves bi'oader, often subcordate, more fleshy, pedicels longer, calyx more inflated, seeds finely granulated, not tuberculate. III. i. Near San Eoque, Wk. 16 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD fVar. longifoUa Wk. has all leaves attenuate at base, linear lanceolate, flowers much larger. III. i. San Eoque, Wk. '■'Saioonaria officinalis L. Sandy places and roadsides ; usually near cottages ; occasional ; 5-7. I. Hedges and waste places in the town, D. III. i. By Lajo in several places ! Lane towards First Venta ! iii. Salt Pans ! Palmones Village ! S. Vaccaria L. Cultivated clayey or sandy fields ; rare ; 4-5. III. iii. Between San Roque and Algeciras, D. \Sagina maritima Don. Sandy or rocky ground near sea; rather rare, or overlooked ; 3-5. Leaves quite obtuse. I. South and west coasts, i?ey.,D. NorthFront! III. ii. Carnero Point, P.L. S. apetala L. Sandy and gravelly fields ; common ; and probably general ; 3-5. Leaves always, and sepals usually, mu- cronate. I. ! III. ii. ! Noted from II., but the plant seen may have been S. maritima. tVar. capillaris Lge. Much more branched, elongate and slender. III. ii. Waterfall Valley ! S. de Palma, Rev. [S. procumbens L. Leaves of what might have been this seen on Glacis and elsewhere, but perhaps only prostrate examples of last.] \Alsi7ie tenuifolia Cr. var. hybrida Wk. Dry, gravelly, and rocky places ; rather rare ; 3-5. Calyx eglandular, leaves out- curved, not strict, taller and less diffusely branched than type, which is not Spanish. I. Mediterranean Steps ! Green's Lodge ! III. i. Alcadeza Crags ! Cork Woods ! A. procumhens Fenzl. Dry rocky places; locally frequent; 3-6. I. Buena Vista to Mediterranean Stops ! Engineer Road ! Eosia ! Above Alameda ! iMoehringia pentandra Gay. Damp woods and rocky water- courses; locally common ; 2-5. Very near M. trinervia Clairv., and perhaps a variety, but sepals shorter than capsule, petals none, stamens about 5 (3-8 in mine), leaves shorter and broader. The nerves vary from three to five in both. III. ii. Waterfall Valley ! Arenaria spatJmlata Desf. Damp clayey spots in fields ■ paths; frequent; 2-5. III. i. and ii. ! A. leptoclados Guss. Dry rocky slopes; rather rare; ^ Kelaart's record of A. serpyllifolia L., of which many autij consider it a slender variety, doubtless belongs here. Perez Lar- thinks all he has seen belong here. I. Near Farringdon's. South and west slopes, K. III. i. North slopes Alcadeza Crags ! A. emarginata Brot. Dry sandy plains ; common ; 2-4. A dwarf inconspicuous annual, with white or rose notched petals. III. i. and ii. ! \A. algarbiensis Welw. Similar places? rare; 3-4. A dwarf very slender annual, leaves very narrow, short, petals rather large, white, entire. III. ii. Near Algeciras, Porta & Eigo. A. montana L. Bushy or rocky places on mountains ; rare ; 4-6. Flowers large, white, leaves lanceolate, glabrous. I. K. Unconfirmed, but quite a likely habitat. III. ii. Summit of El Frayle ! A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 17 Moenchia octandra Gay. Damp grassy and partially shady places ; common ; 2-4. III. i and ii. ! Eeaches top of Queen of Spain's Chair ! Cerastium glomeratum Thuill. {G. viscosum L.). Eoadsides, waste ground, &c., very common ; 2-5. Varies much in habit, often apetalous. I. ! II. ! III. ! jC. brachypetalum'DeBp. Dry gravelly places ; rare? 3. More slender, pedicels mucli longer, usually deflexed in fruit, claws of petals glabrous, or ciliate only at base, capsule one-third longer than calyx. I. Mediterranean Steps ! III. i. Path to First Pine Wood ! Both teste F. N. Williams. C. Boissieri Gren. Eocks and debris on mountains; the type, with densely white-tomentose leaves, rai'e; 4-5. I. Near Green's Lodge ! fVar. gihraltaricum Gren., leaves green, but viscid, is com- mon. I. ! Stellaria media Vill. In all situations ; very common ; 12-5. I. ! II. ! III. ! S. neglecta Weihe var. umhrosa Opiz. Shady places; probably frequent; 2-5. Elongate, flowers rather large, sepals and pedicels glabrous, rarely slightly pubescent, fruit acutely tuberculate, stamens normally 10, but often 3-5. The type has pubescent pedicels and sepals, and may occur. III. i. Malaga Gardens ! iii. Salt Pans ! Both teste E. S. Marshall. S. uligiJiosaMnrr. Swampy places; rare; 3-5. III. i. Near summit of Queen of Spain's Chair ! Spergula arvensis L. Sandy and gravelly places ; the type rare ; 1-5. III. ii. Algeciras, Wk. fVar. glutinosa Lge. is canescent and glandular ; very com- mon. III. ! Spergularia rubra Pers. var. longipes Lge. Sandy fields ; common ; 1-6. Suberect, straggling, little or not glandular, internodes rather long, stipules narrow, inflorescence lax, leafy, flowers deep rose, seeds not winged. III. ! fVar. innguis Fenzl. {Lepigonum neglectuvi Kind.). Stouter, leaves thicker, cymes glandular, flowers and seeds larger. From the description and synonymy this is surely a var. of S. marina Wk. I. Sands at Gibraltar, K. III. ii. Cornfields at Algeciras, Rev. ^S. purpurea Vex^. [S. diandra Heldr.). Sandy ground; rare? 3-5. Erect, slender, little branched, glabrous except cymes, leaves setaceous, short, flowers large, deep rose, stamens 2-3. II. D. III. ii. Marsh at Palmones Playazo ! ? This form is suberect, rather strict, leaves and internodes long, pedicels long, glandular, sepals very acute, glabrous, petals bright rose, stamens 2-5. Mr. G. C. Druce thinks it may belong here. f/S. atheniensis Asch. {S. campestris Wk. ?). Sandy ground; locally common ; 4-5. Prostrate, densely branched, slender and wholly very viscid, leaves vei'y slender, cymes leafless, flowers Journal of Botany, Feb. 1914. [Supplement] c 18 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD small, deep rose. III. ii. By Palmones Railway Bridge ! iii. Los Barrios aviation ground ! S. viarina Wk. Roadsides, often far from the sea; very common ; 3-5. Prostrate, flowers rather small, dull rosy or whitish, in a leafy glandular cyme, some seeds winged. I.! III.! S. media Pers. {Arenaria marginata Fenzl). Salt marshes; occasional ; 3-6. Stout, often suberect, rather like a very large form of last, capsules much larger, pedicels longer, seeds all broadly winged. 11. D. III. i. Mouth of Guadarranque! ii. Pal- mones Playazo ! fVar. ancjustata Clav. (sub S. marginata) has seeds very narrowly winged. III. iii. Guadacorte Marsh I S. rupestris Lebel. Rocks ; rare ; 4-6. Perennial, prostrate and densely matted, internodes short, stipules large and long, flowers rather pale rose. I. By Buffadero Gate ! S. fimhriata Boiss. Sandy ground ; rare ? 2-3. Its descrip- tion hardly differentiates it from last. I. or II. Sea sand at Gibraltar, Salzm. III. ii. Algeciras, Bev. LINACE.E. \Linum galliaim L. Rough bushy slopes; rather rare; 4-6. I. Lagasca. III. i. Queen of Spain's Chair ! L. setaceum Brot. Similar places ; frequent ; 5-6. I. West slopes, Boiss., D. III. i. Queen of Spain's Chair ! Path to First Pine Wood ! Alcadeza Crags ! ii. Hills near El Cobre ! Carnero Hills ! L. strictum L. Dry gravelly places and old walls ; common ; 4-5. I.! II.! III.! Var. spicatum Pers. (var. axillare G. & G.). Flowers axillary, in a narrow raceme. Much rarer. I. Pourr. III. i. Slopes of San Roque, Boiss. By Lajo above Almendral. L. temie Desf. Dry fields and banks ; very common ; 4-12. I. K. III. ! \L. maritimum L. Maritime rocks ; rare ; 6-7. Much like last, leaves broader, flowers larger, sepals obovate acute, not lanceolate acuminate. I. Grassy places and rocks at South Point, Pourr., D. Kelaart indicates slopes over Engineer Road, below Michael's Cave, as the station for this. L. aJigustifoUumHuds. Sandy grassy places; abundant; 1-5. I. Cumberland Flank ! Rock Gun I Signal Station ! III. ! ■\L. dentmbens Desi. Grassy places ; rare; 4-5. III. i. South slopes of San Roque, D. [L. tisitatissimwn L. Kelaart found this in an uncultivated part of the Rock, but I suspect it was a casual. I have seen it as such in the Dockyard.] Badiola Unoides Gmel. Watercourses and damp banks ; locally frequent ; 4-6. III. i. Cork Woods above San Roque, Boiss. Duke of Kent's Farm, K. Queen of Spain's Chair! ii. Waterfall Valley up to 1000 ft. ! ii. or iii. Sea sand near Palmones River, Bev. a flora op gibraltar and the neighbourhood 19 Malvace^. Malva hispanica L. Dry sandy and stony places ; very common ; 4-7. I. ! III. ! M. sylvestris L. Waste places and roadsides ; rather frequent on Kock, much less so in Spain ; 3-6. Epicalyx narrow, quite free. I. ! III. i. Cachon ! Almoraima Station ! M. nicaensis All. Similar places ; common ; 3-6. Flowers very variable in size. Known from M. sylvestris by broader epicalyx and smaller paler rose flowers, and from M. parviflora by more deeply lobed leaves, larger flowers, and calyx half covering fruit. Its free epicalyx distinguishes it from Lavatera cretica. I.! III.! [M. rotundifolia L. is recorded for Gibraltar by Kelaart and Dautez, and is common according to Debeaux, but I have never seen it. Perez Lara says it is subalpine in S. Spain, and doubts many of the records. Probably M. parviflora has been mistaken for it.] M. parviflora L. Waste ground ; common, at least on Eock; 3-6. Carpels glabrous, with flat rugose back, and very acute edges, corolla very small, hardly longer than calyx. I.! II.! III., I believe frequent, but only noted from Punta Mala! i Lavatera arborea L. Eough ground near sea; rare; more frequent formerly? 3-6. I. Eocks and grassy places, also near Signal Station, K., D. A single plant at Monkey's Cave ! L. cretica L. Eough and waste places by roadsides and buildings; very common; 4-6. Epicalyx broad, connate at base. I.! II.! III.! L. Olbia L. Woods and bushy places ; rare ; 5-6. Shrub, flowers clustered, axillary, leaves with a long terminal lobe. III. Near Gibraltar, Lagasca. Var. hispida G. & G. is more hirsute, especially peduncles and calyx, leaves green above, not canescent both sides. III. i. Near Almoraima, Laguna. ii. By Lobo in Carnero Hills ! [L. maritima Gouan is recorded as abundant on the Eock by Kelaart and Dautez. It does not grow in the neighbour- hood. I cannot suggest what species can have been taken for it.] L. triloba L. Bushy ground; rare; 5-6. A tall shrub, flowers clustered, leaves cordate, orbicular, obsoletely 3 -lobed. Winkler named it L. micans L., but Perez Lara thinks it belongs here. III. ii. Algeciras, Winkl. L. trimestris L. Sandy fields ; very common ; 4-6. Erect or prostrate, flowers sometimes white. III. ! AlthcBa officinalis L. Fresh or salt marshes; locally frequent; 6-7. III. iii. Guadacorte Marshes ! Both sides of Salt Pans ! [AURANTIACE^.] [None of the order mentioned by Kelaart is native.] c 2 20 a flora op gibraltar and the neighbourhood Hypericine^. [Hyperimim hircinum L. is recorded by Kelaart as an intro- duced species in Gibraltar, Lem !] H. perfoliatum L. [H. ciliatum Lamk.). Bushy places and banks ; occasional ; 5-6. III. i. Hills over San Eoque Station ! Corkwoods! ii.! iii. Salt Pans! H. perforatum L. var. angustifolium Gaud. Dry bushy places ; common ; 5-7. The type does not occur in the province. I. Above Bruce's Farm ! III. ! H. undulatum Schousb. var. hcsticum Lge. Marshy places; locally frequent ; 7-8. Not seen in flower ; the foliage strongly recalls H. tetrapteriim Fr. III. i. By Lajo below First Pine Wood ! ii. Mountains ! iii. Guadacorte Marshes ! H. tomentosum L. Dry banks and fields ; common ; 6-7. I do not observe much variation in our species ; the two varieties cited differ but slightly from type. III. i. and ii. ! Var. p2ibescens P. L. More erect and woolly-villous, not canescent, flowers larger, sepals narrower and more longly acuminate. III. i. Near San Eoque, Boiss. Var. lusitanicum P. L. Leaves much smaller, hispid, sepals linear-lanceolate, mucronate. III. ii. Algeciras, Rev. jH. cris'pum L. Eough fields ; rare ; 6-7. III. i. Near First Eiver Ferry ! Carteian Hills near Puente Mayorga ! H. humifusum L. Damp grassy fields and watercourses ; occasional ; 5-7. According to Willkomm our form is var. australe Wk., stouter than type, with glandular-dentate sepals, but all I have seen appear to be type. III. i. Queen of Spain's Chair ! ii. Mountains ! [Meliace^.] [Melia Azedarach L. is only planted.] Ampelide^. ■■'•Vitis viiiifera L. Bushy places, hedges, and by watercourses ; frequently quite naturalised ; 5-6. I only give stations where it is established. III. i. Plentiful in Cork Wood Sotos I By Lajo below First Pine Wood ! First Venta, &c. ! ii. Waterfall Valley ! Carnero Hills ! iii. Guadacorte ! Geraniace^. Geranium dissecUim L. Eoadsides, hedge-banks, and grassy places ; rather frequent : 2-5. I. ! II. ! III. ! G. molle L. Similar places ; common ; 2-5. I. ! II. ! III. ! \G.cokimbinuvilj. Woods and shady places; occasional; 4-5. I. Near Signal Station ! III. i. About Almoraima and Long Stables ! ii. Waterfall Valley, from El Cobre to top ; Mountains over Pelayo ! M. de la Torre ! Carnero Hills ! G. rotundifoUum L. Eoadsides and waste places ; frequent, especially on the Eock ; 2-5. I. ! III. i. Cachon ! Puente Mayorga! About San Eoque! First Pine Wood! Malaga Gardens! Almoraima! ii. About Algeciras ! iii. Near Guadacorte ! A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 21 G. Rohertianum L. var. parviflornm Viv. Shady and bushy places ; very common ; 3-5. Differs from type, which is found at Tarifa, chiefly in smaller flowers. I. ! III. ! Erodium ciciUarium L'H6rit. Sandy and gravelly fields and heaths ; very common ; 12-5. I cannot satisfactorily identify all the forms of this aggregate, which, as such, occurs everywhere. I.! II.! III.! E. lyriimdaccum Lge. Similar places ; very common ; 12-5. Our earliest and commonest species, fairly distinct, with several large bright pink flowers, often colouring considerable tracts of ground. III. 1 E. Jacquinianum Fisch. & Mey. Similar places ; 3-6. Densely pubescent and glandular, leaves shortly densely villous, tripinnatisect, segments small, obtuse, peduncles 2-4-flowered, petals subequal, about as long as calyx, not spotted, carpels without a fold, which the last two have. I. D. III. i. Sandy shores by Fort San Felipe, D. Var. bipinnatum Pari. Subglabrous, leaves less deeply divided, segments narrower, petals unequal. I. Brouss. E. Salzmanni Del. Similar places ; 3-6. Stout, stems dark purple, leaves bipinnatisect, segments narrow, acute, peduncles 6-10-flowered, stout, densely glandular, petals equal, scarcely longer than calyx, not spotted, carpels with a long beak, fold small or none. I. Near Old Mole, D. II. D. III. i. Bond's Farm ! ? Andalucian racecourse ! ? Both stout forms with con- siderably divided leaves, and very long carpel beaks. E. moschatum L'H6rit. Grassy and waste places ; abundant ; 1-5. I. ! III. i. and ii. ! Doubtless occurs everywhere. E. malacoides Willd. Eoadsides, banks and bushy places ; very common ; 2-6. Varies considerably in leaf lobing, and somewhat in colour of flowers. I.! II.! III. i.! Less frequent beyond San Eoque. ii.! Kather rare, not seen in mountains. Var. subtrilobum Lge. Leaves tri- or pinnati-lobed, terminal lobe largest, occurs frequently with type. I. ! II. ! III. i. I E. lacmiatmn Willd. Similar places; rare; 4-6. Inter- mediate between last and next, nearer the latter, but more slender, sepals much shorter, 2-2| lines instead of 5-6, beak shorter, l|-2 in., and more slender, carpels without a ridge. I. K. Europa Point ! ? rather near last. Var. involucratum Wk. is stouter, with larger bracts. II. Wk. I Willkomm's specimen looks very like E. Botrys, but differs in calyx, carpel and beak, and its lower leaves are more entire. E. Botrys Bertol. Sandy and gravelly places ; very common ; 3-5. Stems long, straggling, flowers large ; dark reddish, often striped deep red, carpels with 2-3 deep furrows, beak 3-4 in. long. [I. \Vk.'] Willkomm's No. 556 labelled " at Gibraltar and between Chiclana and Conil " is certainly E. laciniatum. III. i. and ii. ! OXALIDE.E. Oxalis corniculata L. Dry banks, path-sides, and by streams ; rather frequent ; 3-6. I. Below Devil's Gap ! Buena Vista ! 22 A FLORA OP^ GIBRALTAR AND THE NEIGHBOURHOOD Below Lunatic Asylum ! III. i. Cork Woods ! Lajo above San Roque. Arroyo Viejo ! Queen of Spain's Chair! ii. Mountains! Algeciras Station and by Miel beyond it ! Guadalmeci I ■•'0. cernua Thunb. Bushy banks, roadsides, and watercourses; very abundant ; 12-4. This species has increased enormously in recent years. I. ! III. ! Zygophyllace^. Tribulus terrestris L. Sandy ground; local; 5-7. I. North Front ! II. ! Not " in great abundance," as Kelaart says. III. iii. Palmones, Bev. RUTACE^. Buta montana Vill. Dry bushy hills ; locally rather frequent ; 6-7. Leaf segments narrow, flowers bright yellow. III. i. Carteian Hills near San Roque ! Towards First Pine Wood 1 Pinar de los Bigotes ! Near Alcadeza Crags ! B. clialepensis L. Dry bushy hills ; locally common ; 2-6. 1. 1 A form at Levant closely approaches next ! III. i. Queen of Spain's Chair ! Cork Woods and elsew^iere ! ii. Thinly scattered on hills and mountains ! B. bracteosa DC. Similar places ; common ; 2-6. Differs in much broader bracts, and shorter petal fimbriaB. I. ? K., who does not record B. chalepensis, which he may have mistaken for it. III. i. Carteian Hills ! Path to First Pine Wood 1 Magazine Hill ! ii. Carnero Hills ! CORIARI^. Coriaria viyrtifolia L. Thickets in marshes and by streams ; locally common ; 3-6. III. i. Arroyo Viejo and Lajo in many places! Railside beyond Almoraima ! iii. Guadacorte Marshes! Rhamnace.e. Bhavmus Alaternus L. Dry bushy slopes and woods ; rather frequent ; 12-3. I. ! A narrow-leaved form near Signal Station may be the form near B. integrifolia mentioned by Kelaart ! III. i. Queen of Spain's Chair ! Cork Woods ! ii. Mountains to highest ridge ! Carnero Hills ! B. oleoides L. Rocky bushy places ; frequent ; 3-5. Leaves oval or elliptical, often small. I. ! III. i. San Roque, Wk. Queen of Spain's Chair ! Cork Woods ! ii. S. de Palma, Wk. iii. Palmones village, Laguna. Var. angustifolia Lge. has linear lanceolate leaves, like those of next, but more rounded on sides. I. Near Windmill Hill Road, Wk., &c. III. i. S. Carbonera, Wk. B. lycioides L. ? Similar places ; rare ; 4-6. Leaves narrow linear, or strap-shaped. Perez Lara thinks the records for the province refer to the last var. I. Communicated, K. III. i. and ii. Precipices at San Roque and S. de Palma, K. B. Frangula L. Wooded valleys ; locally frequent ; 4-6. III. ii. Waterfall Valley ! A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 23 Var. longifolia Eouy is a larger bush or small tree, leaves up to 5 in. long loy 3^ in. wide, fruit larger, more obovate, seeds black, not yellow. Grows with type, into which it runs. III. ii. ! Ilicine^. Ilex Aquifolmm L. Mountain woods ; locally frequent ; 4-5. III. ii. From Waterfall to highest ridge, very small bushes, with less spiny leaves than usual, not seen in flower ! Var. halearica Wk. (I. Perado Ait.). Leaves much larger, usually quite entire. III. ii. S. de Palma, Bev. Terebinthace^. Pistacia Terehinthus L. Woods and rocky slopes ; occasional ; 4-5. I. ! III. i. South slopes of San Roque, D. P. Lentiscus L. Similar places ; very common on Eock and in III. i. I. ! III. i. ! ii. Palmones Playazo ! Occasional in mountains! iii. Guadacorte ! Palmones Sands ! [Schinus molle L. is only planted.] Leguminifer^. AnagT/ris foetida L. Rough bushy hills ; rather frequent ; 12-3. I. Old Man's Garden! Above Wilhs's ! Below Signal Station ! Alameda Gardens ! III. i. San Roque, Wk. Almoraima Station ! ii. Hills north and west of Algeciras ! Carnero Hills ! iii. Near Los Barrios Station ! Retavia monosperma Boiss. Sandy ground ; rare ; 12-2. [I. Planted.] III. i. Linea ! Probably Dautez's S. Carbonera Station. Spavtium junccum L. Rough bushy places ; local ; 4-6. I. Engineer Road ! Jews' Cemetery to Gymnasium ! III. i. San Roque, D. ii. Very common at iVlgeciras, D. I have never seen it there. Sarothanmus bceticus Webb. Woods and bushy places ; locally frequent ; 1-4. Leaves large, trifoliolate. I. Levi. ! Bambur ! Whole side of Rock below Michael's Cave covered with it, K. It is not so now. III. i. Slopes of San Roque, Boiss. d Bent. ! Cork Woods ! ii. Mountains ! Carnero Hills ! S. Wekvitschii B. & R. Woods ; locally frequent ; 4-6. Like S. scoparius, but pod densely hairy, leaves few or none. III. ii. Waterfall Valley ! S. grandiflorus Webb ? Bushy places ; rare or error ? 4-6. Hardly distinguishable from last, and recorded by no other collector. I. Bambur. Cytisus tribracteolatus Webb. Wooded slopes ; locally frequent ; 1-3. III. ii. Waterfall Valley and other valleys north of it ! G. candicans DC. Woods; type rare? 1-5. Fruit straight, rufescent-hirsute. I. Wk., K. Upper Alameda, Frere. Var. Kunzeanus P. L. differs in several minor characters, fruit curved, white-woolly, but apparently becomes rufescent in herbaria, so that the best distinction is lost. All I have seen has pod hairs 24 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD quite white. It is very abundant. III. i. Cork Woods ! ii. Mountains ! Carnero Hills ! Palmones Playazo ! iii. Guada- corte ! Palmones, Lag^ma. C. linif alius Lamk. Bushy slopes and woods ; locally common; 3-6. I.! III. i. Corkwoods! ii. Mountains! Carnero Point! C. triflonis L'Herit. Woods ; locally frequent ; 1-4. Flowers pendulous, rather pale greenish yellow, on long virgate branches. I. Clus., Tournef. Not there now, I think. III. i. Cork Woods ! ii. Mountains ! Carnero Point I Calycotovie villosa Link. (" Spartium villosum L.," Kel.). Dry hills and bushy slopes ; abundant ; 12-5. I. ! III. ! Genista scorpioides Spach. Kough bushy places ; rare ? 3-5. Very near next, but bracts at base of pedicels ovate, not subulate, spines stouter and longer. III. ii. S. de Palma, Bev. G. triacanthos Brot. Similar places ; locally frequent ; 3-8. A rigid dark green bush, flowers usually in short subtruncate racemes. III. i. Queen of Spain's Chair ! Cork Woods ! ii. Palmones Pinar ! Mountains ! Var. galioides Spach has stouter spines, longer leaves (3-4 lines), and denser racemes. III. ii. Hills above Algeciras, Ball. S. de Luna, Laguna. |G. Winkleri Lge. Bushy hills; rare; 3-5. Like next, but leaves mostly trifoliolate, not simple, and bracts 3-4 times as long as, instead of little longer than pedicels. III. i. San Eoque, Winkl. G. gihraltarica DC. Similar places ; locallj' common ; 4-7. Flowers usually in long tapering racemes. [I. Included by Kelaart in error (see K. Fl. p. 93).] III. i. Queen of Spain's Chair ! Cork Wood Crags ! Between Almoraima and Long Stables ! ii. Mountains ! \G. Scorpius DC? Kocky hills and ravines; rare; 2-7. A stout-spined shrub with the aspect of a Cahjcotome. Pods much longer than last. Can Laguna have mistaken C. villosa for it ? I. West slopes, Laguna. j:G. Hmnseleri Boiss. Bushy hills; rare; 5. Very distinct, with a look of Spartium junceum, but flowers smaller. I. K. III. i. South slopes of San Roque, D. Pterospartum lasianthum Wk. Heaths and rocky places on mountains ; rare ? 12-6. Differs little from next, chiefly in more woolly-tomentose, not silvery-silky calyx. Willkomm thinks they should perhaps be united. III. i. S. Carbonera, D. Almoraima, P. L. ii. Mountains, Webb, &c. S. de Luna, Laguna. S. de Saladillo, P. L. P. tridentatum Wk. Similar places ; locally common ; 12-6. III. i. Queen of Spain's Chair ! Cork Woods ! Pine Wood Plains ! Alcadeza Crags 1 ii. Mountains ! M. de la Torre ! Ulex. ^StauracantJius has calyx as long as keel, lips shortly united at base, upper divided to middle. ^Nepa has calyx about half as long as keel, lips similarly united, upper quite shortly bidentate. ^Eu-ulex has calyx lips quite free, as long as keel. This section has the habit of U. europmus. A FLORA OF GIBRALTAR ANB THE NEIGHBOURHOOD 25 U. apkyllus Link. {^Stauracanthns). Woods and heaths; frequent; 2-4. III. i. Cork Woods! ii. Heath near Palmones Pinar ! S. de Palma, Clem. U. spartioides Wk. {^Stauracanthus). Similar places; rare? 3-4. I cannot distinguish this from last. III. i. Pine woods on south slopes of San Eoque, D. There are no pine woods there now. ii. S. de Palma, Clem., Webb, Wk., &c. U. Boivini Wehb {iNepa). Similar places ; frequent or com- mon ; 5-11. Teeth of upper calyx lip deltoid. III. i. Queen of Spain's Chair ! Almorairaa, Laguna ! ii. Frequent in mountains ! Var. megalorites Ball. Calyx teeth elongate, sometimes one- third as long as lip. III. i. Near San Eoque, Ball. S. Carbonera, D. ii. S. de Palma, Bev. \U. luridus Wk. {^NejKt). Very like last var., from which Webb's description hardly differentiates it. III. ii. Near Algeciras, Porta d Bigo. U. scaber Kunze {%E^i-ulex). Bushy places; rare? 12-4. Spines slender, rather pliable, liowers rather small. Abundant elsewhere in the province. III. ii. Waterfall Valley ! fVar. glabrescens Webb has calyx hardly puberulous, with broader teeth. III. ii. S. de Palma, Wk., Bev. iU. br achy acanthus Boiss. [^EiL-idex). I know nothing of this beyond Boissier's description. III. ii. Near Algeciras, Porta ct Bigo. U. aiLstralis Clem. {^Eii-ulex). {U.parviflorusVo\x\^x.). Similar places ; common ; 2-4. III. i. Queen of Spain's Chair ! Carteian Hills ! ii. Mountains ! [U. bceticus Boiss. {lEu-ulex). Kelaart says this covers many acres near Campamento. No Ulex is common there now, the very similar U. australis occurs sparingly.] Adenocarpus grandiflorus Boiss. Wooded slopes ; locally frequent ; 4-6. III. i. S. Carbonera, D. Cork Woods ! Alcadeza Crags ! ii. Mountains ! [Lupinus albus L, is an occasional escape at Algeciras.] L. hirsutus L. Sandy ground; occasional; 4-5. II. K., D. III. i. By Campo Common ! Below S. Carbonera, -D. Almoraimal ii. Woods near El Cobre ! L. angustifolius L. Sandy ground ; frequent, at least locally ; 3-5. III. i. Spanish Eacecourse, Wk. Cork Woods ! ii. Pal- mones Pinar and Playazo ! Behind Sandy Bay ! M. de la Torre ! iii. Palmones Sands ! L. luteus L. Similar places ; rather frequent ; 4-5. III. i. San Eoque Station ! Lajo Valley, below San Eoque ! Near Almoraima ! Foot of S. Carbonera, K., D. ii. Palmones Pinar ! Cachon Farm ! iii. Near Salt Pans ! Ononis campestris Koch & Ziz. Eough gravelly or clayey places; rare? 6-8. Stem hairy in lines, leaves narrow. I. Talbot. 10. repens L. var. horrida Lge. Similar places; occasional; 6-8. Stem hairy all round, spiny. III. i. Carteian Hills ! Near Francia's Farm ! Path to First Pine Wood ! Malaga Gardens ! ii. Carnero Hills ! 26 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 0. plnnata Brot. Sandy ground ; locally frequent ; 6-7. III. i. Cork Woods ! ? leaves only. ii. Palmones Playazo ! (Pro- bably Schott's " near Algeciras "). iii. Palmones Sands ! 0. Picardi Boiss. Loose sand ; locally abundant ; 4-7. Annual, slender, diffuse, flowers rosy, tip of keel bright yellow. II. D. III. i. Pine Wood Plains ! Alcadeza ! Foot of S. Carbonera, D. ii. S. de Palma, Bev. iii. Palmones Sands ! Guadacorte ! 0. Cossoniana B. & R. Light sandy soil ; occasional ; 4-6. Much stouter and more erect than last, more leafy and with larger flowers, 5-6 lines long. I. Sentry Fence on North Front ! II. III. i. Andalucian racecourse ! ii. Behind Sandy Bay ! fVar. rotundifoUa Wk. has larger rounder leaflets. I. Rev. A weed in Alameda Gardens ! Dockyard ! III. i. Near San Roque, Bev. 0. diffusa Ten. Dry sandy ground ; occasional ? 5-6. Differs from last in smaller size and foliaceous floral bracts, but not easily separable. I. Governor's Cottage {K., as 0. serrata Forsk., probably this but not synonymous, see Deb. Fl. p. 59). II. Sea sands on Neutral Ground near Catalan Bay, Boiss., K., D. III. i. Between Gibraltar and San Roque, Salzm. ii. About Algeciras, Bev. 0. mitissiina L. Damp grassy banks and stream beds ; frequent ; 5-6. Annual, much branched, suberect, flowers very small, close set, rosy, bracts broad, white, with strong green nerves at base. I. East side, K. Signal Station Road ! II. K. III. i. S. Carbonera, D. Ditches at Campamento ! ii. By Railway ! Carnero Hills ! 0. filicaulis Salzm. Dry gravelly hills ; frequent ; 5-6. Much like 0. Picardi, but grows in different soil, much less glandular and more greyish-hirsute, racemes more elongate in flower, standard brighter rose, wings and keel white, no yellow tip, and broader calyx segments, not with long subulate points. III. i. Bonel's Farm ! Queen of Spain's Chair ! Alcadeza Crags ! ii. Carnero Hills ! 0. Salzmanniana B. & R. Grassy places and ditches ; rather frequent ; 5-6. A stout suberect annual, spikes long, dense, thick, upper leaves trifoliolate, terminal leaflet very large, and bracts trifoliolate. III. i. By the Cachon ! Railside beyond Almoraima ! By Lajo below First Pine Wood ! Road between Campamento and San Roque ! By the Lobo ! iii. By the Aguacorte ! 0. alopecuroides L. Similar places ; rare ; 5-6. Resembles last, and perhaps a variety. Larger, leaves and bracts all unifoliolate, stipules larger and spike comose. III. i. About San Roque, Brouss. 0. TournefortiiGoss.? Sea sands; rare or error? 4-5. Near next, but much larger and stouter ; leaves trifoliolate, corolla as long as calyx. A conspicuous plant, not recorded by other botanists. II. D. 0. variegata L. Sea sands ; locally common ; 4-5. A small annual, half-buried in sand, flowers bright yellow, twice as long as calyx, and unifoliolate leaves. I. By Sentry Fence ! II.! III. i. Linea ! ii. Sandy Bay ! iii. Palmones Sands ! A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 27 i Var. erioclada DC. I. Sands about Gibraltar, Durand. 0. reclinata L. var. minor Moris. Sandy and gravelly places ; rare; 4-5. Flowers rose, soon pendulous, on long peduncles. The var. has shorter corolla and pods as long as calyx. The leaves in all I have collected are broadly oval or suborbicular. I. Signal Station ! Princess Amelia's Battery ! Above Mediter- ranean Tunnels ! II. Ji., D. 0. yendida Desf. Grassy or rocky places ; occasional ; 3-5. Habit of last, but flowers much larger, with yellowish, not rosy or whitish, wings. I. Buffadero Gate to Monkey's Cave ! III. ii. Near Upper x\queduct! Towards Saladillo ! Carnero Hills! Sandy Bay ! 0. imhescens L. Dry stony or sandy soil, especially dry water- courses; frequent; 4-6. Flowers large, yellow, peduncles muti- cous. I. Above Alameda Parade! Above WilUs's ! About Michael's Cave, K. D., Bev. 0. viscosa L. Similar places ; frequent ; 4-6. Like last, but taller and much laxer, stems often red, peduncles longer and awned. I. Michael's Cave ! Spur Battery ! Levant ! Above Engineer Eoad ! III. i. and ii. ! Var. foetida Deb. has narrower leaves, larger flowers and longer pods, about twice as long as calyx. III. i. Alcadeza Crags ! ? By Cagancha below San Eoque ! ? Both these are perhaps as near type as var. ii. Algeciras, Rev., Winkl. j; 0. crotalarioldes Goss. Similar places ; rare ; 5-6. Like last, but all leaves unifoliolate and pods much inflated, much longer than calyx. III. ii. Foot of S. de Palma at Algeciras, Winkl. fVar. ruhricaulis Wk. Elongate, with purple virgate branches and smaller leaves. III. ii. Near Algeciras, Hegelm. f 0, breviflora DC. Eocky grassy slopes ; rare ; 5-6. Like 0. viscosa, and often regarded as a variety, but corolla shorter than calyx, and other differences. III. i. South slopes of San Eoque, D. 0. ramosissima Desf. var. gracilis G. & G. (0. gibraltarica Boiss.). This species and the next vary indefinitely, and it is difiicult to fix on any characters by which to separate them. I have regarded this as a dark green shrublet with very short glandular pubescence, and narrow elongate leaves. All other points seem unstable. The var. differs in its more slender virgate habit, with smaller leaves and peduncles with longer awns. 1. Catalan Bay ! Casual plants at Haynes's Cave and in Dock- yard ! III. i. Linea ! Near Eocadillo ! iii. Palmones Sands ! 0. Natrix L. var. media Boiss. Dry hills and sandy ground ; occasional ; 4-8. As I understand it, a pale grey-green shrublet, with very long viscid pubescence and broadish leaflets, racemes more compact, the whole plant with the aspect of a large 0. puhescens. It does not appear to be confined to sand dunes. The var. has smaller leaflets, often folded, and peduncles longer than floral leaves. These characters bring it very near 0. ramo- sissima. I. Boiss. ? ex Deb. Fl. p. 63. Boissier does not record it from the Eock in Voy. Bot. K., D. III. i. Linea, K., Fl. p. 178, not Neutral Ground, as cited by Debeaux. 28 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD fVar. inceqiiifolia Mut. Some, often very few, leaves pinnate, the alternate pairs of pinnae small and close to the leaflets ahove them. III. i. Cork Wood Crags ! Path to First Pine Wood ! Malaga Gardens ! Medicago lupulina L. Rather damp grassy or shady places ; rather frequent ; 4-5. Spikes longly peduncled, of twelve or more very small flowers and small smooth fruits. II. ! III. i. Ahout Almoraima and Long Stables ! By Lajo above Almendral ! ii. Near railway about San Bernabe ! iii. Gardens at Salt Pans ! [M. sativa L. as a cultivation escape has been found on the Eock and Neutral Ground.] M. marina L. Sand dunes ; rather frequent ; 3-6. Densely white-tomentose. I. Beyond Catalan Bay ! North Front ! II ! III. i. Near Pedrera ! ii. Palmones Playazo ! Sandy Bay ! Towards Frayle Point ! iii. Palmones Sands ! ill. orbicularis All. Grassy places ; occasional ; 4-5. Fruit discoid, f in. in diameter over thin spineless wing. I. Below Eock Gun ! Grassy ravines, K., D. III. i. About Long Stables and Almoraima ! Near Malaga Gardens ! Near mouth of Guadar- ranque, K., D. ii. Algeciras, Winhl. iii. Hills near Los Barrios Station ! M. intertexta All. Grassy places ; rather rare ; 4-5. Large, hirsute, fruit very large, globose, closely embraced by long inter- twining glabrous spines. III. ii. Near Algeciras, Winkl., Fritze. Below railway near San Bernabe ! iii. Hills near Los Barrios Station ! Salt Pans near Palmones Village. M. ciliaris Willd. Dry hills ; rare ; 4-5. Near last, but fruit grey-green from hispidity of spines, which are much shorter. III. i. At Carteia ! \M. obscura Eetz. var. tornata f. inermis Urb. Waste sandy or grassy places ; occasional ; 3-6. Facial veins running into a lateral nerve, not a marginal one. I have only found the form cited which has an unarmed pod of four or more turns with thin edges, easily mistaken for unarmed forms of M. Uttoralis. I. Ee- clamation Eoad ! Dockyard ! Behind Assembly Eooms ! M. truncatula Gaertn. var. longiaculeata Urb. {M. tribuloides Desr.). Similar places; locally frequent; 3-6. Facial veins as last, but fruit subcylindrical, pubescent, spines stout from a very swollen base, not grooved nor hooked, usually parallel to axis and intercrossing. The var. cited is, according to Willkomm, the only one known in Spain, tilf. uncinata Willd., found by Kelaart at Gibraltar, is a form with uncinate spines. I. From Buena Vista to Governor's Cottage ! Eeclamation Eoad ! M. rigidula Desr. [M. Gerardi Kit.). Similar places; rare; 3-6. Facial veins running into a marginal band instead of a lateral nerve, spines stout, though less so than in M. truncakda, and radial, not axial. It somewhat resembles M. nigra, but has stouter spines, neither grooved nor uncinate. I. Eeclamation Eoad ! M. Uttoralis Eohde. Usually in deep sand ; very common ; 3- 6. Veins as last, stems very long, straggling, fruit usually with A FLORA OP GIBRALTAU AND THE NEIGHBOURHOOD 29 few irregular radial spines, often curved and uncinate, varying much in length. fVar. inermis Moris. Spines none or reduced to tubercles. Known from M. tornata by flat faces and thick edge of spirals. II. ! Var. hrcviseta DC. Spines equal to, or shorter than, diameter of spiral. A common form. I. Reclamation Road ! II. ! Var. longiseta DC. Spines longer than diameter of spiral. The commonest form. I. ! II. ! III. ! -\ M. turbinata V^iWd. Waste sandy and grassy places; rare? 3-6. Very pubescent, pods rather large, tub-shaped, spines reduced to tubercles. M. tuherculata Willd. is very near this, but has aristate, not muticous, peduncles, and glabrous pods flat at base. III. i. At Carteia ! By Lajo near upper ford ! Var. aculeata Gaertn. [M. olivceformis Guss.) has a spiny pod. I. Sandy fields near the Neutral Ground, K. III. ii. Algeci- ras, Wk. M. Murex Willd. var. sj^hcerocajya Bertol. Similar places ; rare ? 3-6. Even nearer last than M. tuherculata, but pod glabrous, rounded at both ends. II. Sandy fields on Neutral Ground at foot of Fort San Felipe, D. III. ii. Algeciras, Fritze, Bev. Near San Bernabe, very glabrous, with short radiating spines ! M. hisinda Gaertn. Similar places ; common ; 3-6. Spirals lax, hardly contiguous, spines radiating, grooved, and usually uncinate. Var. denticulata Urb. {M. denticulata Willd.) has only 1-3 turns of spiral. III. iii. Near Los Barrios Station ! Var. ijentacycla Urb. [M. nigra Willd.) has 5-6 turns and long spines. One of the largest and most conspicuous species. M. Terebellum Willd. differs only in its shorter spines, and M. lap- jKicea DC, which may occur, in fewer (3-4) turns. I. K., as M. Terebellum. Reclamation Road ! Below Mediterranean Road 1 II. K., D. III. ! M. coronata Desr. Similar places; rare? 3-6. Spirals as last, but fruit very small, of one or two turns, with short pectinate spines parallel to axis. I. K. M. minima Lamk. Similar places ; frequent or common ; 3-6. Much smaller, fruit globose, with long tine spines radiating spherically. I. ! III. ! Melilotus parviflora Desf. Fields and waste places ; frequent ; 3-6. Flowers usually very small, fruit globose, very coarsely reticulate. Large-flowered forms look very like M. elegans. I. ! III. i. Slopes of San Roque, D. ii. Railside near Algeciras ! Near Reina Cristina, quite prostrate ! \M. alba Desr. Rough slopes; rare; 4-5. III. ii. By the Lobo! \M. elegans Salzm. Fields and waste places ; frequent ; 3-5. Fruit ellipsoid, transversely rugose. Seen in more stations than those noted. I. Sandpits ! Reclamation Road ! East slopes below Middle Hill ! III. ii. Marsh near Palmones Pinar ! M. sulcata Desf. Similar places ; very common ; 3-5. Fruit with fine regular concentric ribs. I. ! III. i. and ii. ! 30 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD fVar. angustifoUa Willd. with small flowers and narrow leaves, much resembling M. iMTviflora. I. Eeclamation Eoad ! Dockyard ! Var. major Camb., much larger and stouter, with broader leaves, and larger flowers. I. North Front, Boiss. II. A prostrate form ! III. i. At foot of San Eoque, Boiss. ii. Near Eeina Cristina ! El Saladillo ! \M. infesta Guss. Similar places; rare; 4-5. Fruit ribs fewer, wider, much less regular. 1. Winkl. tilf. messanensis Desf. Wet grassy places and marsh borders ; locally common ; 3-5. Eacemes short, axillary, very shortly peduncled, fruit very large. III. i. Punta Mala ! III. ii. Marsh at Palmones Playazo ! iii. Salt Pans ! \ Tr if oU um filiforme Ij. Grassy places ; rare; 4-6. Like next, but heads fewer flowered and laxer, pedicels longer than calyx tube, petiolule of terminal leaflet not or little longer than those of lateral. III. ii. By upper aqueduct ! T. minus Sm. Grassy places; occasional; 4-6. Flowers smaller than next, standard smooth, keeled, wings not diverging. More general than my records show. III. i. Bonel's Farm ! Campamento Common ! Alcadeza Eavine ! iii. Guadarranque Marshes ! T. 2^rocumbens Sm. Dry bushy or stony places; very common; 4-6. Standard flat, strongly striate, spoon-shaped at tip, wings diverging. Very variable. Not synonymous with T. agrariuvi L. as Debeaux supposes. I. ! II. ! III. ! Var. majus Koch {T. campestre Schreb.). Erect, heads large and bright yellow, usually retaining their colour when dry. I. Moorish Wall ! Var. minus Koch {T. 2)}'ocumbens Schreb.). Often decumbent, flowers paler, drying whitish, peduncles shorter. Usually regarded as type. I. Signal Station ! III. i. Wooded slopes of San Eoque, Boiss. T. repcns L. Damp grassy places and by watercourses ; rather common ; 4-6. I. Eoadsides and ditches, K., D. III. ! Eeaches the mountains ! T. isthnocarimm Brot. Grassy places ; locally frequent ; 4-6. Flowers dull rather deep rose, pedicelled, bracteolate. II. Munhy\ III. ! chiefly in ii. ! +T. Jaminianum Boiss. Similar places ; rare ; 4-6. Very near last, if not synonymous ; a specimen at Kew from the original station at Algiers shows hardly any difference. Flowers white, calyx segments narrow, unequal, longer than tube, free part of stipules longer. III. ii. Algeciras, Porta d- Bigo. T. nigrescens Viv. Dry grassy places ; locally abundant ; 4-6. Like T. repcns, but annual, not creeping, heads smaller, on axillary, not long pseudo-radical peduncles ; corolla white, not pink as described. I. ! II. ! III. i. From North Front to Campamento ! T. glomeratum L. Eough banks and grassy places ; rather rare; 4-6. Heads globose, sessile in nearly every axil. Known from T. scahrum by rose, not white flowers, bracteolate, not quite sessile, and pubescence not scabrous. III. i. West slopes Queen A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 31 of Spain's Chaii- ! Eoad to Bonel's Farm ! Alcadeza Crags I ii. Valley north of Waterfall ! fT. cerninim Brot. Damp grassy places; very rare; 4-5. Heads small, lax, shortly peduncled, flowers white, soon deflexed, calyx teeth long, rather slender, spreading, longer than corolla, leaf-teeth setaceous, terminal seta long. III. iii. A single plant in Palmones Sands ! \T. suffocatuvi L. Dry grassy places; rare; 3-5. Heads very small, sub-basal in axils of long petioles. I. On debris of forts, D., Rev. Path below Breakneck Battery ! T. strictum L. Grassy places ; rare ; 5- 6. Erect, leaves narrow, bright glossy green, denticulations glandular, stipules very broad, corolla rose. III. i. Near ford to Pine Wood Plains ! ii. Near iVlgeciras, Bev. Valley near Saladillo ! T. resujnnatiovi L. Grassy places ; very common ; 8-6. Annual, heads small, corolla bright rose, inverted, calyx inflated and woolly in fruit, upper teeth much larger than lower, project- ing from the wool. I. Europa Flats ! II. ! III. ! Occasionally wdth white flowers ! T. tomentosuvi L. Similar places ; equally common ; 3-6. Like last, but corolla normal, fruit calyx quite hidden in wool. I. Below Victoria Battery ! Near Michael's Cave ! Eock Gun ! North Front ! II. ! ? III. ! T. fragiferum L. Similar places ; less common ; 5-7. Peren- nial, flowering much later, flowers much paler pink, heads larger, calyx less woolly, upper teeth about equal lower, slightly project- ing. I. K. III. i. Kailside beyond San Eoque! By Lajo in many places ! ii. Carnero and i\lgeciras Hills ! Near El Cobre ! iii. Palmones Sands ! Guadacorte Marsh ! T. striatum L. Similar places; rare; 4-6. Standard and wrings connate with stamens into a long tube. III. ii. Near Algeciras, Bev. \T. 'pratense L. ? Similar places; rare, or error; 5-6. I. K. III. i. San Eoque, D. ii. Algeciras, D. T. baticum Boiss. Woods and bushy places ; locally rather frequent ; 4-6. Heads large, flowers white or pale cream, rarely pinkish. III. i. Almorairaa ! Path from San Eoque to First Pine Wood ! Alcadeza Crags ! ii. Beyond El Cobre ! Carnero Hills ! T. angustifolium L. Dry grassy places ; common ; 4-5. Heads very long, cylindrical, leaflets long and narrow. I. ! III. ! T. stellatum L. Eough waste or grassy places ; very common ; 3-6. Heads subglobose, calyx segments stellately spreading in fruit. I. ! III. ! T. lappaceum L. Dry grassy places ; frequent or common ; 4-5. Heads peduncled, calyx teeth little plumose, leaf sheaths long and strongly nerved. III. ! T. Cherleri L. Similar places ; common ; 4-5. Heads sessile, very plumose, with very broad involucrating stipules. III. ! T. albidum Eetz. [T. squarrosum DC, T. panormitanum Presl.). Damp grassy places and roadsides ; occasional ; 4-5. Stout, 32 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD rigid, with rather large heads of yellowish- or greenish-white flowers on axillary peduncles, calyx segments indurated in fruit, lower tooth very long and strongly deflexed, stipules very long. III. i. Eoadside near San Eoque ! Near First Venta ! Path to S. Lorca ! ii. Path to El Cobre ! Watercourses below El Saladillo ! Carnero Hills ! iii. Near Guadacorte ! T. leucanthum M. Bieb. Similar places ; rare ; 5-6. Allied to last, but much smaller, heads very longly. peduncled, mostly terminal. III. ii. Near Algeciras, Porta d: Bigo. T. maritimum Huds. Damp grassy places, often remote from sea ; common ; 4-6. Glabrous, heads terminal, ovoid, flowers dull rosy, calyx closed at throat. II. ! III.! especially in ii. ! \T. Juliani Batt. {T. Xatardi DC.'?). Similar places; rare; 4-6. Probably only a variety of last with subequal calyx teeth. III. ii. Near Algeciras, Porta ct Bigo. T. ligusticum Balb. Shady places; rare; 4-6. Eesembles T. lappaceum, but with cylindrical, not globose heads, occasionally twin, and shorter, broader, less strongly nerved leaf sheaths. III. ii. In one or two valleys in the mountains ! T. arvense L. Dry sandy fields and hills ; rather frequent ; 4-6. Usually erect, slender, pale green, with rather small ovoid white plumose peduncled heads. II. D. III. i. Campamento Common ! Near Soto Gordo ! Field beyond Almoraima ! Lajo Valley ! ii. Palmones Pinar ! iii. Palmones Sands ! About Guadacorte ! T. Bocconi Savi. Dry rough places ; rare ; 4-6. Heads small, few, sessile, axillary and terminal, often twin, with pale flowers, calyx open at throat, lowest tooth twice as long as others. III. i. Alcadeza Crags ! ii. Near Algeciras, Bev. T. scabnwi L. Eoadsides and dry places ; abundant ; 3-6 ; Usually prostrate, with sessile axillary heads of small white flowers, calyx closed, much indurated and nerved in fruit, with rigid divaricate teeth. I. ! II. ! III. ! iT. subten-aneitmLi. Grassy places ; common; 2-6. I. Europa Flats ! II. ! III. ! \Anthyllis Vulneraria L. Dry banks and grassy places ; locally frequent ; 3-5. The type with pale yellow corolla and concolor- ous calyx is rare. II. D. Var. ruhriflora DC. Calyx wholly or partly purple, corolla yellow or red. I have only seen it yellow. I. Mediterranean Steps and slopes below ! III. i. Everywhere on slopes of San Eoque and Fort San Felipe, K., D. ii. S. de Palma, K., D. Carnero Point, and scattered to hill tops ! A. cytisoides L. Dry bushy slopes; locally frequent; 4-6; Shrub, 4-6 ft. high, like a Cytisus, flowers in rather long dense racemes. III. i. Cork Wood Crags ! Physanthyllis tetraphylla Boiss. Eough slopes and grassy, sandy places ; common ; 3-5. I. Chiefly south ! III. Eeaching high up mountains ! Cornicina Laflingii Boiss. Uncultivated fields ; rare ; 5-6. Differs from next in much longer peduncles, lateral leaflets few, A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 33 smaller than terminal, pod annulate. III. ii. Algeciras, Nee, Webb. C. liaviosa Boiss. In deep sand ; locally common ; 4-5. I. Catalan Bay ! III. i. Pine Wood Plains near Lajo ! Cork Wood Crags ! ii. Palmones Playazo ! iii. Palmones Sands ! Dorycnopsis Gerardi Boiss. Bushy and heathy places ; rather rare ; 5-6. III. i. Queen of Spain's Chair ! By Lajo at Pine Wood Plains ! Alcadeza Kavine ! ii. By Miel below El Cobre ! iii. Palmones Village, Rev. Bonjeania recta Eeichb. Marshy places and by rivers ; frequent; 5-6. III. ! Very abundant in marsh beyond Almoraima! B. hirsiUa Eeichb. Dry bushy slopes ; rather rare ; 4-6. III. i. Magazine Hill ! Long Stables Eavine ! Path to First Pine Wood ! Near Alcadeza ! Tetragonolobus j^urjmretis Moench. Open slopes and grassy fields, especially cultivated ; frequent; 3-4. I. Chiefly north-west slopes ! Windmill Hill, K. Near Inundation, K. III. i. and ii.! \T. fseuclo - purpureus Uechtr. Similar places; rare; 3-4. Differs in shorter pod with shorter and broader wings. Porta & Eigo's specimen at Kew looks like a mere variety of last. My 519 from Middle Hill may be this. III. i. Field borders near San Eoque, Porta iC Bigo. ii. Meadows near Algeciras, Fritze, Winhl. \T. conjugatus Ser. Similar places; rare; 3-4. Flowers smaller and paler, pod much more slender, not winged. III. ii. Meadows near Algeciras, Fritze, Winhl. \T. siliquosus Eoth. Similar places ; rare ; 5-6. Perennial, with large solitary yellow flowers on long peduncles, pod narrowly winged. I. Lag. \Lotus edulis L. Dry sandy or grassy slopes ; common ; 3-5. Flowers 1-2, pod much inflated. I. ! III. i. and ii. ! Eeaches Carnero Point ! L.ornitlioiJodioides'Li. Similar places; common; 3-5. Flowers several, smaller, with long pendulous pods. I. ! III. ! L. cijtisoides L. Dry rocky and bushy places ; locally abundant ; 3-5. Like L. corniculatns, but in larger laxer patches, often sub- scandent, wings connate in front. A form with pale or whitish yellow flowers occurs. I.! III. i. Euins of Carteia ! Slopes of San Eoque, D. L. creticus L. Sand dunes ; rather frequent ; 2-7. Large, prostrate, leaves silvery-silky, flowers large. L. Salzmanni B. & E. and L. commutatus Guss. are varieties differing chiefly in form and relative lengths of calyx teeth. I. A plant or two at Light- house ! Catalan Bay ! North Front Sentry Fence. II. ! III. ii. Palmones Playazo ! Sandy Bay ! iii. Palmones Sands ! L. corniculatus L. Dry grassy places ; rare ? 3-5. Perennial, differing also from L. arenarius, which often closely resembles it by its entire styles. I. Signal Station Eoad ! A very large form like L. uliginosus, which the Eev. E. S. Marshall thinks near var. crassifolius Pers. This may be the form mentioned by Kelaart in Fl. p. 98, as near L. major Scop. III. i. Beyond Alcadeza Crags ! Cork Woods ! Journal of Botany, March, 1914. [Supplement] d 34 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD Var. hiisutus Koch is a more hairy variety. I. Dry grassy slopes in south and west, K., D. iL. uliginosus Schk. Damp bushy places ; rare ; 4-6. Much taller and stouter, tips of calyx teeth divergent in bud. III. ii. By streamlet above El Cobre ! L. arenarius Brot. Dry sandy places, often in deep sand ; locally common ; 3-7. Annual, with flowers inclining to orange, styles bifid. I. K. ! II. Very abundant ! III. i. Towards Pedrera ! Var. major Wk. By description this differs chiefly in size, but L. canescens Kunze, cited as synonymous, is densely covered with yellowish tomentum and looks very different. I. Jews' Cemetery ! ? A glabrous plant as stout as L. uliginosus, with the bifid style of L. arenarius, and differing from both in its long narrow calyx tube, with teeth much longer than tube. L. angustissimus L. Sandy ground; rare? 4-6. Differs from next in several unstable characters, the most constant being its long slender fruit, 3J-5 times as long as calyx. Young examples placed to L. liispidus may belong here. II. Damp sand, Boiss., D. Ill, ii. Algeciras, ib. L. liispidus Desf. Sandy places, and in woods ; common, at least locally ; 3-6. Fruit not more than 2| times calyx or less, thicker than in last. The corolla often turns green after drying, but this may not occur for months. III. i. Pindalista ! ii. Near Cortijo Trinidad ! M. de la Torre ! Mountains ! iii. Guadacorte and Guadarranque Marshes ! L. ijarviflorus Desf. Dry or wet places on heaths and open mountain slopes; locally common; 3-6. Petals and fruit scarcely longer than calyx. The corolla ultimately often turns green, as in last, but this is not a constant character. I. Behind the Mount, K. Probably the reference on p. 60 to L. angustifolius refers to this. III. i. Bonel's Farm ! Alcadeza ! Pine Woods near San Eoque, Wk. ii. Mountains ! Psoralea bitwninosa L. Eough or bushy places ; common on Eock, much less so in Spain. It varies much in size. I. ! III. i. Alcadeza ! ii. Algeciras ! In the mountains ! M. de la Torre ! Carnero Hills ! Var. latifolia Moris (P. plumosa Echb., P. imlcBstina Mor.) is a large form, which Perez Lara does not distinguish even as a variety. I. ! III. ii. Algeciras, Bev. Galega officinalis L. Bushy places by streams ; rather com- mon ; 4-5. III. i. Lajo Valley below Almendral ! A large mass in field beyond Almorairaa, and elsewhere ! ii. and iii. ! Astragalus epiglottis L. Dry sandy places; rare ; 3-5. Annual, with bicuspidate, centrifixed hairs, flowers very small, whitish or pale yellowish, turning blue, in dense axillary, very shortly peduncled heads. I. Lag., Winkl. III. iii. Palmones, Eev. A. pentaglottis L. Dry gravelly places; occasional, or locally frequent; 4-5. Flowers pinkish purple, in dense round heads, on very long peduncles. III. i. Opposite Francia's Farm ! Carteian Hills ! Between San Eoque and First Pine Wood ! North of San A FLORA OF GIBRALTAR AND THE NEIGHROURHOOD 35 Roque ! ii. Eailside and adjacent fields near Algeciras ! Carnero Hills ! iA. algarbieiisis Goss. Sand dunes ; very local; 4-5. Annual, with yellow flowers in very short racemes on very long axillary peduncles, racemes elongating in fruit, fruit obliquely ovoid, apiculate, glabrous. III. iii. Sandy hillock west of Guadacorto Marsh ! A. hamosus L. Dry gravelly places; rather frequent; 3-5. Like next, but smaller and more diffuse, with falcate pods. I. ! II. D. III. i. Slopes of Fort San Felipe, D. River bed above Almoraima ! ii. Algeciras, Bev. iii. Near Salt Pans ! A. bceticus Li. Similar places ; equally frequent ; 4-5. I. Rock and North Front ! II. ! III. i. About San Roque ! ii. Algeciras, Bev. Carnero Hills ! [A. dej^ressits L., communicated to Kelaart by Lemann, was probably an error, though he is usually accurate. It is not known in the province, nor is it North African.] A. kisitanicus Lamk. {Phaca hcetica L.). Woods and shady places ; rather rare ; 2-4. A large coarse species, with racemes of large white flowers, and much inflated pods. [I. Cultivated ? -ST.] III. i. Spanish racecourse, -D. ; not there now. Sandy fields at San Roque, D., K. Cork Woods ! Biserrula peleciims L. Cultivated and fallow fields ; rare ; 3-6. Pods long, pendulous, scalloped. III. i. Near San Roque, Brouss. Scorpiurus s^dcata L. Cultivated and rough fields ; abundant as an aggregate ; 3-6. The type has spiny pods, the convolutions all in one plane. The floral characters are quite unreliable. I only cite stations where I have collected the varieties. I. Dock- yard ! Europa Glacis ! Var. subvillosa P. L. Pod very similar, but convolutions very irregular and not in one plane. As abundant as type. I. Levant ! Sandpits ! III. i. Beyond Bond's Farm ! ii. Railside near Algeciras ! Var. muricata P. L. Spines very short, or reduced to tubercles. Decidedly rarer. III. i. Campamento Common ! Near S. Lorca ! S. vermiculata L. Similar places ; abundant in Spain ; rare on Rock ; 3-6. Spines cylindrical and densely contiguous, covering the whole pod. I. Rare, K. II. ! III. ! Coronilla glauca L. Bushy places and ravines ; frequent on Rock; 1-3; rare in Spain; 3-5. I. Chiefly about Mediterranean Steps, and above Lower Lines, scattered elsewhere ! III. i. Ravine near Long Stables ! C. juncea L. Similar places ; locally common ; 3-5. III. i. Cork Wood Crags ! [Hippocreins comosa L. is recorded from San Roque by Kelaart only.] ill. unisiliquosa L. Grassy places; rare; 3-5. III. i. Orchard by Arroyo Viejo, below Malaga Gardens ! H. nmltisiliquosa L. Dry places and cultivated soil; rather common ; 3-5. I. ! III. i. and iii. ! d 2 36 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD Ornithoims repandus Poir. Dry gravelly or sandy places ' local or rather rare ; 3-5. Annual, lower leaves simple, rest pinnate, with the lower pair of pinnae large and stipuliform, true stipules minute. III. i. Cork Wood Ci-ags ! Near Second Pine Wood ! Almoraima ! Bonel's Farm ! 0. scorjnoides L. Similar places and in cornfields ; locally common ; 3-5. Closely resembling last, but leaves trifoliolate, lowest pair stipuliform, the terminal very large. III. i. About Pinar de los Bigotes ! 0. ebracteatus Brot. Sandy fields; frequent; easily overlooked; 3-5. III. ! 0. compressus L. Similar places ; very common ; 3-5. I. At foot of Rock, Lem. II. ! III. ! 0.roseuslj.{0.sativus'Qro\>.). Sandyplaces; very common; 3-5. Pods straight, with contiguous joints, beak curved, by description as long as last joint, but I find it usually twice as long. III. ! Var. macrorhynclms P. L., has beak 3-4 times as long as last joint. III. i. Almoraima, Porta & Bigo. Var. isthmocarptis P. L. Pod much curved, joints separated by a distinct isthmus, beak 3-4 times last joint. I. Uncultivated sands at Gibraltar, Brouss. III. ii. Algeciras, Bev. Hedysarum capitatum Desf. Dry gravelly hills ; occasional or locally common ; 4-5. Annual, with deep rose, not crimson flowers. III. i. Opposite Francia's Farm ! Carteian Hills ! Alcadeza Plain ! jff. coronarmm L. Fields and hills ; abundant in Spain ; 4-6. I. Very rare, K. III. ! Certainly not abundant in Cork Woods, as Debeaux reports ; I have not seen it there. H. humile L., var. majus Lge. {H. Fontanesii Boiss.). Bushy places ; rare ; 3-5. Leaves much narrower, stipules connate. III. i. Near San Roque, Boiss. Onohrychis eriopliora Desv. Sandy ground ; rare ; 4-6. Re- sembles Hedysarum coronarium but smaller, with much smaller narrower leaflets, and pod of one woolly joint. III. ii. S. de Palma, B. & B. ii. or iii. Sands by Palmones River, B. £ B. Vicia sativa L. Bushy places ; frequent ; 3-5. A protean species, not always easily separable from V. angustifolia. I. ! II. ! III. ! Var. macrocarpa Moris. Flowers 1 in., pods 2-2|- in. by 3-5 lines, prominently reticulate. Perez Lara says this is the most frequent form, and it is the only one admitted by Debeaux, but I have not identified it. III. Between the Neutral Ground and Algeciras, K., D. i. Slopes of San Roque, ih. Var. cordata P. L. Lower leaves obcordate, upper bilobed, flowers 6-7 lines, pods l|-2 in. by 2-3 lines, reticulate, but not prominently. I think common. I.' Near WiUis's ! Mediter- ranean Steps ! III. ii. S. de Palma, Bev. My 2012 from near El Saladillo has very peculiar leaflets 1 in. by h in., with a long cuspidate apex, not retuse. It resembles nothing I know. V. angiLstifolia Roth. Similar places ; frequent, but less so than last; 3-5. Smaller in all its parts, with narrow more cylindrical pods. III. ! A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 37 Var. segetalis Thuill. Usually regarded as the type, has obcordate or oblong-lanceolate leaflets. III. i. Queen of Spain's Chair ! ii. Palmones Pinar ! Var. Bohartii Forst. Upper leaves or all with narrow linear leaflets. Perez Lara takes this as the type. III. ii. Waterfall ! My 1998 is a very narrow-leafleted form ! V. hyhrkla L. Cultivated or waste places and roadsides ; occasional or rare; 3-5. Closely resembles V. lutea, but standard villous on back, and leaflets usually truncate or retuse. I. Wk, Foot of Mediterranean Steps ! III. i. Foot of San Roque, D. V. lutea L. Similar places ; occasional ; 3-5. Flowers very pale yellowish, rarely slightly washed with violet, standard glabrous, leaflets acute or acuminate, not retuse. Only the var. is recorded by Debeaux. I. Above Willis's ! Sandpits ! III. i. Near Linea Cemetery ! About Almoraima ! Var. hirta Boiss. Much more pilose, with longer denser hairs on pods. III. i. or ii. Cultivated and sandy fields by Guadarranque, D. V. vestita Boiss. Mostly in cultivated fields and railway banks; frequent, 3-5. Flowers not yellow as described, but rather dark purple, wings and back of standard dull yellowish. III. ! fVar. tiiherculata Wk. Pod with large scattered tubercles bearing long white hairs. This only is recorded by Debeaux, though the type is more common. III. i. Marshy places in sand desert, D. ii. Algeciras, Winkl. My 662 from railside may be this. t V. onobrychioides L. Wooded slopes ; rare ; 5-6. A climbing perennial, with long peduncles bearing 6-12 large violet flowers, with paler keels, calyx tube straight, pod long, 5-10-seeded. III. ii. Near Algeciras, Winkl. \V. cassubica L. Similar places; rare; 6-7. Like last but erect, with more numerous flowers on peduncles shorter than leaves, pod short, rhomboid, 2-3-seeded. III. i. About San Eoque, Brouss., Lag., Durand. f V. tenuifolia Roth. Shady ravines ; rare ; 5-6. Perennial, calyx tube obliquely truncate or rounded at base, not saccate. Very like V. Cracca but stouter, peduncles longer than racemes, the whole longer than leaf, flowers pale blue, limb of standard longer than claw. III. i. South slopes of San Roque, D. V. Cracca L. Woods and bushy places ; locally frequent ; 4-7. Flowers shorter, violet blue, standard shorter, limb as long as claw, peduncles shorter than or equalling raceme, the whole rarely longer and often shorter than leaf. III. i. Near Almo- raima! Long Stables Ravine I ii. M. de la Torre! Mountain Valleys 1 A white flowered form occurs ! V. varia Host. Wooded and bushy places ; rare ; 4-6. Near last, but annual or biennial, calyx saccate at base, corolla larger, violet, with wings paler or whitish, standard shorter than claw. III. i. Almoraima, P. L. About San Roque, Boiss. 1 V. villosa Roth. Similar places ; rare ; 5-7. Very near last, but flowers opening in succession from bottom, not all simul- 38 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD taneously, raceme plumose in bud, lower calyx teeth longer and narrower, and corolla wholly blue. III. i. Lajo bed above upper ford ! Alcadeza Crags ! Above Almoraima Station ! ? ii. S. de Palma above Algeciras, Fritzc. V. pseudo-cracca Bertol. var. multiflora P. L. Similar places; rare ; 5-6. Near last. The type has only 3-6 flowers, blue with yellow wings, but the var. with 8-20 blue flowers seems indistin- guishable from V. villosa. III. ii. Garganta del Capitan, P. L. V. atroimrimrea Desf . Dry fields and bushy places ; occasional ; 4-5. I. K. II. A single plant! III. i. Bond's Farm! Pine Wood Plains ! Almoraima ! Carteia, K. ii. Palmones Pinar, &c. ! iii. Guadacorte ! ■f-F. boetica Lge. Eough bushy places; locally common; 5-6. A straggling climber, with pale blue and white flowers. III. ii. Wooded hills above Algeciras, Fritze, Winkl. Near Carnero Point and in Frayle Valley I f F. clisperma DC. Sandy fields ; rare ; 4-5. Peduncles aristate, flowers small, style laterally compressed, bearded under apex only, pod obliquely truncate and beaked at apex. III. ii. Near Algeciras, Bev. V. gracilis Lois. Damp grassy places ; frequent ; 4-5. I. Bushy places, K. III. ! Abundantly in Guadacorte Marshes! V. imhcscens Boiss. Bushy places ; usually in woods ; locally abundant ; 4-5. Peduncles muticous, pod 4-6-seeded. III. i. East slopes Chair ! Long Stables ! ii. M. de la Torre ! Mountains ! [F. Ervilia Willd. is cultivated about San Eoque and Algeciras.] Lathyrus Clymenum L. Bushy places and cornfields ; rather frequent ; 3-5. There are two colour forms ; one has a dark purple standard with paler bluish wings, the other a dull cherry red standard with lavender wings, or moi'e rarely a bluish pink standard and white wings {L. articulatus L. ?). The former is the prevailing form on the Eock and in uncultivated soil, the pale one usually in cornfields. Var. latifolkis Godr. with oblong lanceolate leaflets, and var. temiifolius Godr. with linear lanceolate ones grow together, the former the commoner on the Eock. I ! III. i. and ii. ! The dark form is rarer, that with white wings only on railway near Algeciras ! L. Ochrus L. A weed of cultivation and by roadsides ; frequent ; 3-5. I. D. III. i. and ii. ! L. Apliaca L. Waste and cultivated places, bushy hills and woods ; rather frequent ; 3-6. I. Gardens, even in the Town, D. III. i. About Almoraima and Long Stables, &c. ! ii. and iii. ! L. anntms L. Sandy and gravelly hedgebanks and grassy places; occasional; 4-6. III. i. Campamento Common! Almo- raima! Between Neutral Ground and Guadarranque, D. ii. Seen but no stations noted ! iii. About Salt Pans ! L. Cicera L. Cultivated fields and bushy places ; rare ? 3-5. Flowers crimson or scarlet, peduncles muticous, pods large oblong. Differs from L. setifolius in size, and in style twisted so as to A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 39 appear flattened laterally at apex. III. i. Slopes of San Roque and Queen of Spain's Chair, D. iii. Eailside at Guadacorte ! ■■'•]- L. sativus L. Cultivated fields, also in woods; occasional; 3-6. Flowers turquoise blue and white. III. i. and ii. ! Some- times a field full. Cork Woods about Almoraima, rare ! *Var. sti'imlaccus Wk. (L. quadrimarginatus Bory & Chaub. var. amphicarpos Wk.). Flowers blood red, pods winged back and front. III. Near Gibraltar, Lag. L. hirsut'us L. Damp grassy or marshy spots ; occasional ; 4-6. Flowers dark violet, pods very hispid or villous. III. i. Beyond Almoraima Station! ii. Eailside near San Bernabe, with broad leaflets! Near El Saladillo, with narrow leaflets 1 iii. Guadacorte Marshes ! Salt Pans ! L. tingitaims L. Bushy places ; rare ; 4-6. Flowers large, deep crimson, keel with a long narrow beak. I. K., D. III. i. Between Linea and S. Carbonera, Porta A Bigo ! ii. Algeciras, Rev. By Arroyo Gaba! Eailside towards M. de la Torre ! [L. odoratus L. has been found as an escape at Algeciras by Eeverchon.] L. latifolius L. Bushy places and woods ; frequent ; 5-6. Varies considerably in width of leaflets, but I have not seen good var. angustifolius. III. i. Cork Woods ! Arroyo Viejo ! Alca- deza ! ii. Carnero Hills ! iii. Guadacorte Marshes ! L. angulatus L. Grassy places ; rare ? 3-5. A small annual with purple or violet, not scarlet flowers, peduncles awned, much longer than petioles, pods linear, seeds angular, tuberculate. I. Finlay ! III. i. Andalucian Eacecourse ! Bonel's Farm ! ii. Algeciras, Winhl., Bev. Almoraima, Porta c0 Bigo ! L. sphcericiis Eetz. Similar places ; rare ; 3-5. Very similar to last, but usually larger, flowers scarlet, peduncles as long as or longer than petioles, seeds globose, smooth. III. i. About Almoraima and Long Stables ! L. setifolius L. Eough bushy places ; rare '? 4-5. Flowers scarlet or brick red, peduncles muticous, pods rhomboid. Known from L. Cicerahy its smaller size, and style not twisted. I. Leml Above Willis's ! Above Engineer Eoad ! ? III. ii. About Water- fall ! ? No specimens kept from stations queried. [Pisum arvense L. is an occasional weed in cultivated fields. Seen also at Sandpits.] [Erytlirina Corallodendron Willd., Dolichos Ugnosus L., D. imrpureus L., Acacia Farnesiana Willd., Gleditsia tri- acanthos L., Cercis siliquastrum L., and Ceratonia Siliqua L. are all cultivated species, but the last-mentioned occurs subspon- taneously about San Eoque.] EOSACE^. Bubus ulmifoUus Schott {B, discolor \^. & N.). Bushy places; common, often very abundant ; 4-10. I can only distinguish one fairly constant species, which the Eev. W. Moyle Eogers thinks best under this name. I. Engineer Eoad ! Lower Lines! III. I 40 A FLORA OF GIBKALTAR AND THE NEIGHBOURHOOD Var. amcemis P. L. is regarded as the usual form, common throughout Andalucia, but no specimens I have examined agree with Debeaux's description in Fl. p. 78. III. ii. Algeciras, liev. Bosa sempervirens L. Bushy places ; especially by water- courses ; frequent ; 5-7. III. ! Especially common in Cork Woods ! B. micrantha Sm. Bushy places ; rather rare ; 4-6. The segregate was named B. septicola D6s6gl. for me by Prof. Dingier. III. i. By Lajo near Second Pine Wood ! About Almoraima ! ii. By Miel from near source to below El Cobre ! [B. canina L. recorded by Kelaart is not a native of Gibraltar.] [Amygdalus communis L. occurs subspontaneously, more fre- quently beyond our limits. A tree or two grows some way above luce's !] iSjnraa flabellata Guss. Bushy hills; rare; 5-6. III. i. S. Carbonera, Bev. [Fragaria vesca L. recorded by Kelaart as cultivated in Gibraltar gardens.] \Potentilla Tormentilla Sibth. Bushy places in mountains ; locally frequent ; 3-6. Only var. elatior Lehm. is given by Debeaux. It is taller, leaves with broader segments, and stipules tridentate, but I have seen nothing different from the usual British form. III. ii. 1 P. reptans L. Damp grassy hollows and stream beds ; fre- quent ; 4-7. III. ! Frequently not flowering. Agrimonia Eupatoria L. Damp bushy spots and by streams ; frequent; 4-7. III.! iAlchemilla arvensis Scop. In short grass in rather dry places; frequent; 3-5. I have only seen small forms resembling A.micro- carpa B. & E. in short calyx limb, but the fruit is ovoid, not subglobose. III. i. First Pine Wood! Cork Woods! ii. Moun- tains to Guadalmeci ! Poterkim verrucosum Ehrenb. (P. mauritanicum Boiss.). Dry roadsides and fields ; frequent ; 4-5. Glabrous or densely pilose below, fruit obsoletely tetragonous, densely and coarsely tuber- culate. I. About Willis's, &c. ! Engineer Eoad ! II. ! III. ! Var. Magnolii P. L. Fruit more angular, ribs more visible, but it hardly differs, and all I have gathered fit type best. III. i. South slopes Queen of Spain's Chair and San Eoque, D. ii. S. de Palm a, I). P. omdticaule B. & E. Bushy or heathy places on hills ; locally frequent ; 3-5. Much smaller, usually stemless. III. i. From Queen of Spain's Chair to Cork Wood Crags ! CTatcegus monogyna Jacq. Bushy places and woods ; frequent; 3-5. I. ! III.! Abundant in parts of Cork Woods and at foot of Algeciras mountains, more rarely near summits ! [C. maura L. f. is now regarded as synonymous with last. It is at most a form with narrow cuneate leaves, trilobed at apex only. Good forms seem quite rare, but not confined to Eock.] G. brevispina Kunze. Similar places ; rare ? 3-5. Differs from C. monogyna in little but purple-veined petals, and blood A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 41 red, not bright red fruit, and I suspect it is merely a form of that species. III. i. San Eoque, Wk. ii. Algeciras, Wk. Mr. Druce tells me he has seen it at the latter station. Pyriis communis L. var. Mariana Wk. Bushy hill slopes ; occasional; 2-3. Only this variety is recorded. It differs in spinescent branchlets, ovate or oval-subrotund leaves, and tur- binate globose fruit on very thick peduncles. I have not seen any named specimens ; my own agree very closely with P. cordata Desf. in their small neat leaves, but that species is not recorded for Spain. [I. The type cultivated, if.] III. i. East slopes Chair, very small bushes almost buried in heather ! Alcadeza Crags to Majarambout ! ii. S. de Luna about Algeciras, P. L. iii. Sandy ground near Guadacorte Marsh ! ■''Cydonia vulgaris Pers. is quite naturalized here and there, as beyond Almoraima and Long Stables, remote from buildings, also in Palmones Sands, small barren bushes looking quite native. [Primus Armeniaca L. and Eriobotrya japonica Lindl. are only cultivated.] [Granatace^.] [Punica granatum L., as the remains of cultivation, looks native in several places on the Rock and in Spain.] MYRTACEiE. Alyrtus communis L. Bushy places and woods; frequent or common ; 6-7. III. i. Foot of Chair ! Alcadeza Crags ! Cork Woods ! ii. Palmones Playazo ! iii. Palmones Sands ! Lythrarie^. \Lythrum Salicaria L. Marshes and by rivers ; type rare, var. rather frequent ; 6-9. III. i. Behind Almendral, K. Var. tomentosum DC. More canescent, spikes dense, villous tomentose. III. i. By Lajo in many places! Cork Wood Sotos! ii. Palmones Playazo! Carnero Hills! iii. Guadacorte Marshes, sometimes 6-8 ft. high ! L. Grceffcri Ten. (L. fiexuosiim Lag.). Damp grassy places, roadsides, and ditches; very common ; 4-8. II.! III.! Var. Preslii Deb. Stem erect, simple, lower leaves rounded or subcordate at base, upper narrow. II. and III., K., D. L. Hyssopifolia L. In drier places ; common ; 4-6. Annual, flowers smaller, petals paler, 5-6, also stamens. II. ! III. i. and ii. ! My 1196, from sands near Pedrera, is small, subsimple, 2-3 in. high, and looks different. L. thymifolia L. Dry light soil ; rare ; 5-6. Smaller and more slender, petals very small, 4, stamens 2. III. i. Near San Eoque, Ball ! ii. Railway near Algeciras ! iii. Roadside near Los Barrios Station ! \Peplis Portula L. Marshes and ditches; rare? 5-6. Only the variety is recorded, but my only gathering is type. III. ii. Hills near San Bernabe ! fVar. longedentata J. Gay differs in its sessile flowers. 42 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD bracteoles longer than calyx tube, stipules large, often connate with base of tube, outer calyx teeth twice inner and gland-tipped. III. ii. S. de Palma, Bev. f P. erecta Eeq. Damp sandy places ; rare ; 4-6. Erect, 1-2 in. high. III. iii. Palmones Sands ! Halorage^. jMyriophyllum alterniflorum DC. Pools ; rare ; 4-5. III. ii. Near Palmones Pinar ! Onagrarie^. ■''(Enothera stricta Ledeb. Deep sand ; rather frequent ; 4-6. I. D. II. D. III. i. Linea Sands! Punta Mala! Puente Mayorga ! Near Eocadillo ! iii. Guadacorte ! Salt Pans ! Einlohium hirsutum L. var. villosissivuim Koch. Marshes or by pools; locally frequent; 6-9. III. i. Cork Wood Sotos! iii. Guadacorte Marshes ! Salt Pans I f J5. Tournefortii Michal. {E. virgatum Fr. var. inajus Lge.). Hedges and ditches ; occasional ; 5-8. Allied to E. ohscurum Schreb., but much taller, stouter, darker green, with the look of a glabrous E. parvifloriim. III. i. By the Lajo ! Kailside near San Eoque Station ! ii. Eoadside near Algeciras ! El Cobre ! iii, Guadacorte Marsh ! E. adnatum Griseb. {E. tetracjonum L.) ? Similar places ; rare or error? 6-9. I suspect the last species has been mistaken for this. III. i. or iii. Marshes on banks of Guadarranque, D. Ditches at San Eoque, Poiirr. E. iMTviJiorum Schreb. Similar places ; rare; 6-9. III. ii. At El Cobre, a very white villous form ! ilsnardia palustris L. Springs and running water; rare; 7-8. III. ii. S. de Palma, Bev. Tamariscine^. Tamarix gallica L. Eiver banks and wet places ; rare ? 4-5. Very like next, but racemes lJ-2 in. long, lax and slender, hypogynous disc with ten obtuse angles, filaments salient fi'om them, anthers longly apiculate. [I. Cultivated, K.] II. At foot of Fort San Felipe, D. T. africana Poir. Similar places ; frequent ; 3-4. Eacemes shorter and denser, disc with 5 acute angles, filaments not salient, anthers muticous. [I. By North Front Cemetery, planted !] III. By rivers and at Salt Pans ! Cucurbitace^. Bryonia dioica Jacq. Bushy places, hedges, and woods ; occasional; 3-5. I. K. III. i. Near Fort San Felipe, K., D. About San Eoque and by river below Station ! Cork Woods ! ii. Arroyo Gabo, and elsewhere ! Carnero Hills ! EchalUum Elaterium Eich. Dry rocky d6bris ; locally common ; 1-12. I. From Mediterranean Steps to Buena Vista I Near Michael's Cave ! A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 43 PORTULACACE^. Portulaca oleracea L. A weed in sandy gardens; locally frequent; 2-9. Hardly looks native, but Perez Lara does not question its status. III. i. Linea! ii. Algeciras Station! Near First Venta ! iii. Salt Pans ! Paronychiace^. Corrigiola telephifolia Pourr. var. foliosa P. L. Dry grassy and stony fields ; very common ; 1-12. Between type and C. littoralis. I. K. III. ! \Herniaria incana L. Dry sandy soil ; rare ; 5-6. A whitish pubescent perennial, with very shortly pedicellate flowers, few in clusters or in short axillary racemes, sepals densely pubescent, edges and apex not ciliate. Easily confounded with H. cinerea DC., a frequent Cadiz species, but annual, clusters larger, flowers quite sessile, sepals with longer more rigid hairs and ciliate at apex. II. K., D. III. i. Eailway beyond San Eoque ! ? Per- haps H. cinerea. illlecebrum verticillatum L. Damp sandy places ; rare ; 2-7. III. i. Pools on Bonel's Farm ! The floating form looks very like a Callitriche. ii. S. de Palma, Bev. ChcBtony cilia cymosa Wk. Dry gravelly and sandy places ; occasional ; 4-6. A small slender erect annual. I. Boiss. III. i. Queen of Spain's Chair near Pedrera ! Linea, K. Alca- deza Plain ! ii. or iii. Sands near Palmones, Bev. Paronychia echinata Lamk. Dry sandy, stony or gravelly hills ; rather frequent ; 3-5. III. i. Queen of Spain's Chair to Alcadeza ! iii. Palmones Sands ! P. argentea Lamk. (P. hispanica Clus.). In dry, mostly very sandy places ; abundant ; 1-5. I. ! II. ! III. ! My 285 from Neutral Ground closely resembles this, but has narrow leaves, no petals, and muticous sepals. It may be a distinct species. Lceflingia micrantha B. & E. In deep sand ; very local ; 4-6. A small glutinous, subleafless annual, very like L. hispanica in appearance, but with 5, not 3, stamens, and outer, not all, sepals aristate on either side. III. ii. and iii. Sands near Palmones, on both sides of river ! L. sp. ? (No. 1716). A small slender leafy annual, apparently of this genus, just below cottage at Waterfall ! I have not seen it in flower. Polycarpon tetraphyllum L. f. Dry roadsides, sandy fields and waste places ; abundant ; 3-6. I. ! II. ! III. ! Var. alsinoides Gren. Less branched, flowers fewer, rather large, in small compact cymes. III. iii. Palmones Sands ! ? Var. floribunclmi Wk. Much branched, flowers very numerous, in dense cymes covering whole plant. III. i. Bonel's Farm ! ? Crassulace^. Umbilicus pendulinus DC. Eocks, walls and banks ; very common ; less so on Eock ; 5-6. Cauline leaves often laterally petioled, all crenate. 1. 1 III. ! 44 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD fVar. tmncatns'W .-Dod, in Journ. Bot. 1914, p. 12 ; has all leaves, even radical, laterally petioled, rather deeply lobed. III. iii. Roofs at Palmones ! U. horizontalis DC. Similar places ; common, especially on Rock ; 5-6. Flowers greenish white, ovoid, about 3 lines long, capsules long, lanceolate, acuminate. I. ! III. ! f U. citrinus W.-Dod in Journ. Bot. 1914, p. 12. Sandy banks; rare; 5-6. Much taller, flowers long, cylindrical, yellow, capsules short, linear-oblong. III. i. By Almoraima Soto ! Valley opposite Long Stables ! \U. Winkleri Wk. is now referred to Sedum Winkleri.] Pistorinia Salzmannii Boiss. Sandy ground; locally common ; 5-6. Flowers bright yellow inside, dull reddish outside. III. i. Cork Woods ! By Lajo near Second Pine Wood ! ii. Cortijo Trinidad ! iii. Near Almoraima ! Senipervivum arboreum L. Rocks; locally common; 1-3. I. Bungalow ! From Governor's Cottage to below Mediterranean Road ! iSechtm amplexicaule DC. Dry stony places; rare; 5-7. Inflorescence very lax, flowers distant, radical leaves setaceous. III. i. Neighbourhood of Gibraltar, K. Near San Roque, D. S. altissimum Poir. Similar places ; very common on Rock ; occasional in Spain ; 6-7. I. ! III. i. Cork Woods ! Alcadeza Crags ! ii. Palmones Playazo ! S. acre L. Similar places ; rare; 5-6. III. i. Duke of Kent's Farm, K. S. brevifolium DC. Rocks on mountains ; locally frequent ; 5-6. III. i. Queen of Spain's Chair! Alcadeza Crags ! ii. Moun- tains to summits! i *S'. micrantlmm Bast. Stony places; locally common ; 5-6. I. ! III. ii. Valley north of Waterfall Valley ! Only leaves seen. fS. Winkleri W.-Dod in Journ. Bot. 1914, p. 12 (S. hirsutum var. bcBticum Rouy; Uinbilicus Winkleri \Nk.). Rather damp rocks; locally frequent ; 5-7. Flowers large, white, in a lax raceme, leaves and inflorescence glandular. Near S. hirsutum All, but petals connate in lower part. III. i. Summit of Chair ! Alcadeza Crags ! ii. Mountains to highest ridge ! FlCOIDE^. MesembryantJievmm nodiflorum L. Stony places ; rare ; 5-6. Leaves cylindrical, pustulate, flowers small, whitish. I. Europa Lighthouse and Glacis ! Below Mediterranean Tunnels ! III. ii. Algeciras, D. M. crystallinum L. Similar places ; rare ; 4-6. Leaves flat, ovate, pustulate, flowers large, w4iitish. III. ii. Near Algeciras, Nee. [M. Aitonis Jacq., mentioned by Kelaart as cultivated in Gibraltar, is very near the last. I have seen neither.] [M. acinaciforme L, with very large bright purple flow^ers is extensively planted on forts on the Rock, and about cottages in sand-dunes in Spain, and is becoming naturalized.] A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 45 [M. crassifolium L. ? is planted at Governor's Cottage, and has become quite naturalized on rocks near the Lighthouse.] [Cacteze.] [Opuntia. One or two species are commonly cultivated and grown as hedges, occasionally growing subspontaneously.] Saxipragace^. Saxifraga globulifera Desf. var. gihraltarica Boiss. Eocks, old walls and stony debris ; locally common ; 3-5. I. Northern slopes and precipices from Rock Gun to Castle, and debris slopes below ! Mediterranean Steps ! Umbellifer^. jSa7iicula eurojjcsa L. Shady woods ; rare; 4-6. III. ii. Top of Waterfall Valley ! Eryngium ilicifoliwn Lamk. Sandy ground near sea ; rare ; 6-8. Annual with small heads. A specimen in herb. Balestrino so labelled, without locality, is E. inarUimum, which Kelaart also records. I. North Front, eastern side, rarely, K. II. K. III. i. Linea, K. E. tricuspidatu'm L. Dry fields and heathy places ; occasional ; 6-8. Radical leaves oval cordate, dentate, rarely lobed, cauline with narrow-linear segments, heads small, sessile. III. ii. S. de Palma in Los Barrios district, Bev. E. maritimum L. Sea sand ; occasional ; locally common ; 5-7. I. North Front! II.! III. i. or iii. Near R. Guadarranque, K., D. ii. Sandy Bay ! E. aquifolium Cav. Cultivated or fallow fields; locally common ; 6-7. Radical leaves oblong, subentire, cauline very undulate and spinose-lobate, rarely all subentire. III. i. About San Roque, especially north of it ! ii. About Railway near Algeciras ! E. dilatatum L. Dry heaths and fields ; frequent or common ; 6-8. Radical leaves subpalmately bipinnatisect, lobes lanceolate, incised-spinose, petiole winged and spiny to base, cauline pinna- tifid with broadly linear segments. III. ! Scandix Pecten-Veneris L. Dry cultivated fields and rough ground; frequent; 2-5. I. Levant! Near O'Hara's and Break- neck ! III. i. San Roque, especially north ! Near First Pine Wood! ii. Algeciras Station! El Saladillo ! Near Sandy Bay! Gonopodium capilUfolium Boiss. Dry heathy places ; locally common ; 6-7. Near G. denudatum Koch, but leaf segments longer and narrower, fruit and involucel longer. In my 2079 the involucel is abnormally long, often much exceeding fruit. III. i. By Second Pine Wood ! ii. Slopes beyond Waterfall ! Ammi majus L. Dry fields and roadsides ; frequent ; 5-7. Varies much in height and leaf cutting, from 1-2 to 6 ft., flowers very white, involucre pinnatisect. III. i. About San Roque ! 46 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD ii. About Algeciras, but rare south of it ! iii. Near Los Barrios Station and elsewhere ! Var. glajicifolium G. & G. Stem whitish, leaves bipinnate, glaucous segments linear, entire, pedicels and involucels shorter. III. ii. Near Algeciras, Rev. A. Vis7iaga Lamk. Cultivated fields; locally abundant ; 6-7. Leaves more numerous, very finely divided. III. i. Between San Eoque and S. Lorca ! Koad to Malaga Gardens ! Near First Venta ! Almendral ! River above Almoraima ! iii. A single plant by First Eiver Ferry ! Aijium graveolens L. Marshy places ; rare ; 6-9. I. or II. K., either on North Front or Neutral Ground. III. i. Almoraima Soto ! iii. Salt Pans ! Guadacorte Marshes ! Hdosciadium nodiflorum Koch. Streams and ditches; com- mon ; 4-7. Very variable ; stout, erect, 3-4 ft., or small, decumbent. II. ! III. ! Var. ochreatum DC. is a very dwarf creeping form. II. D. 1 have only seen decumbent forms of type here, much larger than var. III. i. Mouth of Guadarranque, -D. Over PindaHsta! Pivqnnella villosa Schousb. Dry bushy or sandy ground; rather locally frequent ; 7-9. About 1 ft. high, much branched, leafless above, branches deflexed in bud, radical leaves rosetted, bipinnate, with rather large segments, pedicels and petals very villous. III. i. On and below Cork Wood Crags ! iii. A single plant at Palmones Village ! Bidolfia segetum Moris. Cornfields ; very common ; 4-5. Tall, annual, leaf segments capillary, flowers yellow. [I. A casual on Europa Glacis !] III. ! \Petroselinum peregriiiwn Lag. Rough stony places ; rather rare; 4-5. Tall, leafy, flowers greenish yellow, leaf segments broad, fruit laterally compressed, involucre 1-3, involucel several. I. Catchment below Rock Gun ! Buffadero Gate! III. ii. Alge- ciras, Fritze. BuiJleurum protractum Hoffm. & Link. Cultivated fields ; frequent ; 4-6. I. North Front, Frere. III. i. San Roque, especially north ! River bed above Almoraima ! Queen of Spain's Chair ! ii. ! B. paniculatum Brot. Dry heathy and bushy places ; locally frequent ; 6-7. Much branched, very slender, leaves long, narrow lanceolate. III. i. First Pine Wood! Cork Wood Crags and below I \B. folios2im Salzm. Stony slopes and rocks on mountains ; locally common ; 7-8. Stout, simple, leaves close set, lanceolate, longly acuminate, panicle branched. III. ii. Mountains ! B . gibraltaricum L. Rocky places ; rare; 6-8. 2-3 ft. high, leaves mostly radical, twisted so that faces are vertical, involucre and involucel persistent. I. Lem. I Above St. George's Hall, and towards Europa, Wehh, Boiss. A specimen so named from Kelaart is B. fruticosum. B. fruticosum L. Similar places ; locally frequent ; 6-8 Shrub, 4-6 ft. high, leaves scattered, elliptical lanceolate, in- A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 47 volucres very deciduous. I. Signal Station to Breakneck Battery ! Between luce's and Signal Station ! Buena Vista Gorge! Near Farringdon's! Crithmum maritimum L. Maritime rocks; locally common; 6-10. Perez Lara says it is common all round the coast. I have not seen it off the Kock. I. All round, chiefly at Europa ! Kundmannia sicula DO. Eough and grassy hills ; common, locally abundant ; 5-6. Short, with radical rosettes of pinnate leaves, segments broad, flowers yellow, involucres many, reflexed. I. About Willis's! Engineer Eoad ! III.! CEnanthe fistulosa L. Marshes ; locally abundant ; 4-6. Stems soft, flexible, fistular. Ill, iii. Guadacorte Marshes ! QU. globosa L. var. elata P. L. Damp hollows and by streams ; common ; 4-6. Fruit much inflated, in a subglobose head. The type is 6-12 in. high, with 5-6 rays, only 2-3 fertile ; it is not recorded. Var. elata is taller, Sh ft. high, rays 8-15, all fertile. II. ! III. i. and ii. ! fVar. Kunzei Lge. is intermediate in height, with 8-10 rays (the outer barren?), and pedicels thick even in flower. It is the only form hitherto recorded for our region. III. i. Between Gibraltar and San Eoque, Wk., K., D. ii. Algeciras, Rev. QU. pe^tcedanifolia Poll. Damp marshy places, often in clumps of Juncus ; locally common ; 6-8. Leaves all subsimilar, root fibres ovoid, or at least thickened. III. ii. Estuary near Keina Cristina Hotel ! iii. Guadarranque Marshes ! CE. piminnelloides L. Dry places, chiefly in woods; very common ; 4-6. Segments of radical leaves much broader than those of stem ; root fibres with an ovoid or subglobose tuber near end. I. K. III. ! CE. crocata L. By running water ; locally frequent ; 4-6. [II. ? D. Surely CE. globosa has been mistaken for it '?] III. i. Queen of Spain's Chair ! ii. Mountains ! Var. ajnifolia P. L. has a colourless, not ochreous, root juice, and is usually regarded as synonymous. III. i. Between Gibraltar and San Eoque, Wk. Foeniculum officinale All. {Anethum Fceniculum L.). Dry hill slopes and banks ; rare ? 6-8. Perez Lara says it is as common in the province as next, but all I have seen is referable to F. inperitiim. I. K., D. III. ii. Algeciras, Rev. F. piperittim DC. Similar places ; very common ; 6-10. Differs chiefly in its long narrow panicle of sublateral umbels, with much fewer rays, and fewer leaves with shorter segments. I. ! III. ! Magydaris iMnacina DC. Dry slopes ; locally frequent ; 6-7. Tall, white-flowered, leaves all radical, with few very large segments, inflorescence and fruit pubescent. I. From Ince's to Signal Station ! Near Michael's Cave ! III. i. First Pine Wood ! Alcadeza Crags ! Cork Woods ! ii. Waterfall Valley ! Carnero Hills ! Torilis nodosa Gaertn. Eoadsides, and waste places ; com- mon; 4-5. I.! III.! 48 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD T. neglecta Eoem. & Schult. Ditches and hedgebanks; very common ; 4-6. Near T. infesta Hoffm., of which some consider it a variety, but much taller, branched above only, with habit of T. Anthriscus Gmel. T. imrpurea Guss. is a form with purple stem and fruit spines. I. ! III. ! Caucalis leptophylla L. Bushy ground and cultivated fields ; rare? 3-6. Annual, short, much branched, umbels small on short sublateral peduncles, fruit linear oblong, with long spines. I. About the middle part of the Eock, K., Boiss., dc. III. ii. Alge- ciras, Eev. A single plant by railway near aqueduct ! Orlaya iilatycarpos Koch. Sea sand ; rare ; 4-5. Annual, suberect, glabrous, involucres broad, hyaline, flowers white, broadly radiant. I. ? K. Probably Neutral Ground. II. K., D. III. iii. East end of Bay, near the Guadarranque, K., D. 0. maritima Koch. Sea sand ; common ; 4-5. Dwarf, pubescent, involucre narrow, herbaceous, flowers small, pink, hardly radiant. I. Catalan Bay ! North Front ! II. ! III. ! Daucus crinitus Desf. Grassy or sandy fields and hill slopes ; rare; 6-7. Leaves glabrous, segments verticillate. l.Pourr. III. i. Campamento Common! Not in flower, ii. About Algeciras, i?ey. D. Carota L. Grassy or rough hills and banks ; common ; 4-7. Most variable ; umbels flattish in flower, concave in fruit, spines distinct, short or long. D. maritimus Lamk., smaller and more slender, with thicker glabrous, shining leaves, and D. maxi- onus Desf., taller and stouter, with large broad leaves, and larger flowers but smaller fruit, are reduced to varieties by Perez Lara. I. ! II. ! III. ! My 1236 from railway near Algeciras may be D. maximus. D. gummifer Lamk.? Eough places near sea; locally fre- quent ? 5-8. Umbel convex in flower and fruit, branches divaricate, leaves thicker, peduncles stout, fruit spines short, confluent at base into a wing. The Gibraltar plant agrees except in fruit, which does not differ from that of D. Carota. D. gingidium L. is not synonymous, as Kelaart and Debeaux say ; it has a concave umbel and different habit. I. Governor's Cottage ! Mediterranean Steps ! D. vuLricatus L. Field borders and roadsides ; very common ; 4-6. Flowers large, very white, fruit with long slender spines. ILI IIL! Elaoselinum foetiduvi Boiss. Sandy flats and mountain slopes ; common ; 5-6. Tall, flowers yellow, leaves much divided, petioles hispid, involucre 0-1, involucel several, fruit dorsally compressed, with a broad wing, which, as in Thapsia, is not developed till maturity. I. ! III. ! E. Asclepium Bert. {E. meoides Koch., E. temiifolia Lge.). Heaths ; rather rare ; 6-7. Eather tall, slender, involucres 3-4, rays long, 8-12, leaves palmately divided, two lowest lobes short and directed downwards, each lobe rather narrow, tripinnatisect, ultimate segments rather short, narrow and apiculate. III. i. Northern slopes of Chair ! Thajma villosa L. Bushy mountain slopes ; locally common ; A FLORA OP GIBRALTAR AND THE NEIGHHOURHOOD 49 4-7. Much like ElcBoselimim fcetidum, but no involucres, leaves adpressed to ground, with fewer, broader blunter segments. The type has bi- or tripinnatisect leaves, segments short and rather small. I.! III. Mountains! Var. latifolia Boiss. differs greatly in much larger, broader, and fewer leaf segments. III. i. Near Almoraima Station, &c. ! Alcadeza Crags! ii. In and about Palmones Pinar! T. garganica L. var. deciissata DC. Sandy and bushy places ; rather frequent ; 4-8. Leaves very pilose, decussately pinnate, segments broad, decurrent, not all in one plane. I. ? Dry hills at Gibraltar, Nee. Probably in Spain, not on the Eock. III. ! Bifora testiculata Spreng. Cornfields ; rare ; 4-5. Fruit didymous, very rugose. Not synonymous with Coriandrum sativum L. as Debeaux makes it. III. Between Gibraltar and Algeciras, D. i. San Eoque, Campamento, &c., Frere. Capnop]iyllum2:)eregrinumljQ,Ts.\V. Cornfields ; very common ; 4-5. Annual, very glaucous, divaricately branched above, umbels subses- sile, flowers small, white, fruit transversely plicate. III. i. and ii. ! Ferula cotwmmis L. '? Dry bushy slopes ; rare or error ? 3-5. Leaves flaccid, ultimate segments long, narrow linear, entire. I think the next species may have been mistaken for this, though the record is by Lemann, who is usually accurate. I. ? Levi. \F. tingitana L. Similar places ; locally very common ; 3-5. The stoutest and earliest yellow umbellifer, leaves very large, much dissected, subcoriaceous, very shining, ultimate segments oval and lobed, fruit oval, with a thick border. 1. 1 III. ? Near Gibraltar, Fritze, Winkl. This may refer to the Kock itself, I never saw it elsewhere. Opoponax Chironiuml^och. Kough places; rare ; 5-6. Eather like Kimdmannia sicnla, but much larger and more branched, pubescent, rays 20-30, involucres few, fruit dorsally flattened, ribs keeled, margin thick, obtuse. III. ii. Near Algeciras, TFwi^Z. Hippomarathrum Pteroclilcenum Boiss. Sand-dunes ; locally abundant ; 5-6. II. ! III. ! H. BocGoni Boiss. Similar places ; rare ; 5-6. Half the size, fruit smaller, with smooth, not papillose ribs, all involucres entire. Smooth fruits are by no means always associated with entire involucres, they are much divided in Ball's specimen. II. Ball ! III. ii. Near Algeciras, Winkl. Smyrnium Olusatrum L. Waste and bushy places; abundant on Eock, occasional in Spain. I. ! III. i. Malaga Gardens ! iii. About Guadacorte ! Gonium macidatitm L. Waste places and field borders ; rare ; 5-6. I. Burial Ground, K. This may be North Front or Sand- pits. A single plant on North Front I III. i. By river at Almoraima ! ii. Carnero Hills ! Araliace^. Hedera Helix L. Eocks and walls, but chiefly in woods ; locally common; 9-1. I. In gardens or on buildings, but native in many places! III. i. and ii. Abundant in mountains, rare elsewhere!? Journal of Botany, April, 1914. [Supplement] e 50 a flora of gibraltar and the neighbourhood Caprifoliace^. Lonicera implexa Ait, {L. Caprifolium Desf.). Bushy moun- tain slopes; common on Eock, occasional in Spain; 5-6. I. ! III. i. Queen of Spain's Chair ! Alcadeza Crags ! Malaga Gardens ! Cork Woods ! Var. puberiila P. L. Leaves pubescent beneath, corolla glandular, villous or glabrous. III. ii. Slopes beyond Waterfall I L. Periclymemim L. Woods and bushy places ; occasional ; 5-7. Perhaps more frequent than my stations show. [I. Introduced, 7i.] Var. hispanica Ball. Leaves softly pubescent both sides, pedicels, calyx and corolla densely glandular. III. i. Cork Woods ! ii. S. de Palma, Bev. iii. Guadacorte Marshes ! Sambucus nigra L. Woods and near cottages ; occasional ; 4-6, [I. In hedges and gardens, -K".] III. i. Cork Wood Sotos ! Above Almoraima ! About San Eoque and towards Station ! ii. By Miel, &c., about Algeciras ! El Cobre ! M. de la Torre ! S. Ehulus L. Waste places near habitations ; rare ; 6-7. III. ii. Above Miel Bridge ! Below Palmones new bridge ! Vihurnum Tinus L, Woods ; locally frequent ; 3-4. III. i. Queen of Spain's Chair, leaves only ! ? Cork Woods ! ii. Upper mountain slopes ! Eubiace^. Sherardia arveiisis L. Grassy places, fields and woods ; abundant ; 2-6. I. ! II. ! III. ! With pure white flowers near Algeciras ! Asperula arvensis L. Cornfields and open places ; rare ; 4-5. Annual, flowers deep blue, involucred by glabrous leaves. Ill, i. Long Stables ! Slopes of San Eoque, D. ii. South of Algeciras ! A. hirsuta Desf. Stony slopes and fields; rare; 4-5. Peren- nial, flowers rose, panicle leaves much shorter, glabrous. I. K. Ill, i. Slopes of San Eoque, D. Pinar del Eey, Porta d Rigo ! Cr2icianeUa maritima L. Sand-dunes ; rather frequent ; 4-7. Perennial, rigid, scabrous, leaves pungent, short, broad, decussate, flowers yellowish in broad bracteate spikes. I. Catalan Bay ! II. K., D. III. i. Near Tunares! ii. Palmones Playazo! Sandy Bay ! iii. Palmones Sands ! C. angnstifolia L. Dry waste places ; rare ; 5-6. Annual, slender, divaricately branched, leaves short, adpressed, spikes slender. III. i. By Fort San Felipe, D. Slopes of San Eoque, D. ii. Eailway near Algeciras ! Buhia peregrina L. Bushy places ; very common ; 4-6. The type has leaves broadly lanceolate, acuminate. I. ! III. ! Var. latifolia G. & G. (var. lucida Webb). Leaves ovate or obovate-elliptical, shortly and abruptly acuminate. I. Juss., K., D. III. i. Slopes of San Eoque, Boiss. Alcadeza Crags ! ii. Mountains, with leaves up to f in. wide ! Var. angnstifolia G. & G. Leaves linear lanceolate, gradually acuminate. I. Juss., K.,D. III. i. San Eoque, ei. ii. Mountains, the commoner form ! A very peculiar state (my no. 2241) grows in Palmones Sands, doubtless induced by exposure and absence A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 51 of material on which to climb. It is dwarf, erect, compact, and little scabrous. [Galium gihraltaricum Schott fil. A doubtful species, des- cribed in Deb. Fl. p. 95. Perez Lara thinks it may have been a form of G. cavipestre.] G. ellipticum Willd. Shady valleys and mountain slopes ; locally frequent ; 5-6. III. i. Woods near San Roque, Potcrr. Almoi'aima, Reut. ii. Mountains ! [G. rubioides L. Kelaart was informed that this grew in Gibraltar, but it has not been confirmed.] G. paliLstre L. var. elongatum G. & G. Marshes and ditches; occasional or locally frequent ; 5-7. III. i. Cork Wood Sotos ! Lajo Marshes! ii. Shore marshes near Algeciras, Bev. iii. Guadacorte Marshes ! G. campestre Schousb. Cultivated and fallow fields; frequent; 5-6. I. P.L. III. i. Campamento Common! Alcadeza Crags ! Around San Roque ! ii. Carnero and Algeciras Hills ! iii. Salt Pans ! [G. viscosum Vahl. {G. glomeratum Desf.) is recorded by Kelaart from the Spanish Racecourse, doubtless in error for the last.] G. divaricatum Lamk. Dry grassy or stony places ; rare ; 5-6. Very near next, but panicle broader, laxer, branches elon- gate and filiform, leaves not reflexed. I. Pavon. III. i. San Roque, Re2it., D. ii. Algeciras, i?eu. Upper Waterfall Valley ! ? Perhaps G. parisiense. G. parisiense L. {G. anglicum Huds.). Similar places ; rather rare; 5-6. III. ii. Algeciras Station! Between El Cobre and El Saladillo ! Var. vestitum G. & G. has a hispid fruit. III. i. Campo Common ! Railside beyond San Roque ! ii. Railside beyond Algeciras ! G. Aparine L. Roadsides, gardens and hedges ; frequent ; 3-5. I. Chiefly in gardens ! Reclamation Road ! III. ! G. tricorne With. Cornfields ; rare '? 4-5. I. K. III. i. Near Pindalista ! Almoraima ! G. saccharatum All. Dry hills, banks, walls, &c. ; very abun- dant ; 1-4. Large specimens not in fruit resemble G. Aparine. L! IIL! G. murale All. Walls, rocks and dry places ; very common ; 1-5. I. ! III. i. and ii. ! Vaillantia hispida L. Rocks and old walls ; rare ? 4-5. Hispid, inflorescence dense, no horn on top of fruit. I. Near Levant Battery ! W. muralis L. Similar places; locally very common; 2-5. I. ! Sometimes shghtly hispid. My 300, from Catalan Bay, is a very laxly branched form simulating Polycarpon tetryphyllum. Valerianae. Valeriana tuberosa L. ? Stony or grassy places ; rare cr error ? 4-5. Forms of this often resemble C. Calcitrapa, which Kelaart may have mistaken for it. I. South and west slopes, K. e 2 52 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD '■"^Centrantlms ruber DC. Eocky places, probably not native, but quite naturalised; 4-6. I. Above Devil's Gap, flowers usually white ! Parson's Lodge ! C. Calcitrapa DC. Eocks, old walls and stony places ; fre- quent? 4-6. Very near next, but usually shorter and more slender, cymes more compact, and corolla tube only as long as achene. I. About Willis's ! III. i. Cork Wood Crags ! ii. Pal- mones Playazo ! I think it occurs in many other places, but I had difficulty in distinguishing it from next. C. viacrosipJion Boiss. Similar places, and in deep sand ; much commoner; 3-7. Usually much stouter, corolla tube 3 times as long as achene. I. ! III. ! Fedia Langei Pom. Cultivated fields, &c. ; abundant ; 1-5. Known from next by oblong-linear fruit. The relative lengths of corolla tube are not reliable, and the species are indistinguishable in flower. I believe both are equally common I. ! II. ! III. ! F. graciliflora Fisch. k Mey. {F. CormtcojncB Gaertn. ?). Similar places ; abundant ; 1-5. Fruit suborbicular, more or less inflated. Perez Lara gives the synonymy cited, but Pomel shows that there are difl"erences {Vide Deb. Fl. pp. 98-99). 1. ! III. ! Valerianella microcarpa Lois. Dry fields and grassy places ; rather rare ; 2-4. A slender annual, fruit the smallest of our species, less than \ line long, calyx limb very small, obtuse. III. i. Path above Bonel's Farm ! West slopes of Queen of Spain's Chair ! Cork Woods ! V. truncata Betcke. Similar places ; rare ; 3-5. Shorter than last, corymb closer, fruit rather larger, calyx limb reticulate, apiculate, as long and as broad as fruit. III. i. Eiverside above Almoraima ! t V. carinata Lois. Similar places ; rare ; 3-5. Flowers in dense globose heads, fruit If by 1;^ lines, prominently ribbed, crown hardly distinct from body. III. i. By Soto Gordo ! V. coronataT>C Sandy cultivated or fallow fields ; rare; 4-5. A stout annual, very like next, but calyx campanulate, glabrous within. Kelaart wrongly makes them synonymous. III. i. Near San Eoque, Wk. ii. Near Algeciras, Rev. V. discoklea Lois. Similar places ; frequent ; 4-5. Calyx subrotate, tomentose within. I. Moorish Wall! Near Michael's! Engineer Eoad ! III. i. Near Campamento Cemetery ! San Eoque, especially on north side ! Cork Wood Ci'ags ! ii. Near Algeciras Cemetery ! iii. Near Guadacorte ! DIPSACE.E. Dipsacus sylvestris Mill. Damp grassy places ; common ; 6-7. Central scales of inflorescence often long, forming a coma as in D. ferox. III. ! \CepJialaria syriaca Schrad. Dry fields; I'are, probably casual; 6-7. Annual, 6-18 in. high, heads small non-radiant, pale blue or lilac, pales and involucre longly mucronate. III. i. San Eoque, K. Perez Lara wrongly cites the station as Gibraltar. A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 53 Pterocephalus Broussonetii Coult. Dry bushy hills and sandy places ; frequent ; 5-6. Tall, with smaller blue heads than either of the Scabiosce. II. Deb. III. i. Cork Woods! Alcadeza and elsewhere ! ii. M. de la Torre ! Hills north of Algeciras ! iii. Near Los Barrios Station ! About Salt Pans ! Scabiosa maritima L. var. grandiflora Boiss. Dry hills and bushy places ; very common ; 5-6. The type, not recorded, has heads of medium size, corolla lilac rose, or yellowish, ovate- cylindrical in fruit. Var. grandiflora has flowers and heads twice the size, cylindro-conical in fruit, lower leaves serrate, not deeply incised. I. ! III. ! Var. atroimrpurea Boiss. Flowers dark or blackish purple. The flowers vary in shade, but I have never seen any approach- ing the atropurjMrea of horticulturists. I. Eare, K. III. ii. Sands at Algeciras, Nilss. S. stellata L. Similar places ; rather common, especially in III. ii, ; 5-7. I. Eare? III. i. By Lajo! Alcadeza Crags, &c. ! ii. ! iii. Occasional ! Pycnocomon rutcefoliuvi Hoffm. & Link. var. bcsticiim Lge. (Scab, urceolata Desf. var. bipinnatisecta Boiss.). Sand-dunes ; very common ; 5-8. The variety is taller, 2-4 ft., heads twice the size and more radiant, and phyllaries connate to one-third, instead of half their length. II. 1 III. ! Composite. Eupatorium cannabinum L. Marshy ground; locally fre- quent; 6-8. III. i. Cork Wood Sotos ! Bellis annua L. Open grassy places ; abundant ; 12-5. Small simple forms are var. minuta DC. A form with lavender, not at all pink or red ray florets, is abundant on Neutral Ground and by Devil's Tower. I. Behind the Grand Stand and by Devil's Tower! Debris at foot of Forts, K. II. ! III. ! B. perennis L. Similar place ; rare ? 2-5. Closely resembles small states of B. sylvestris, but its disc achenes are glabrous on the edge, not ciliate. The faces in both are puberulous. I. ? K. III. i. Cork Woods near Almoraima ! B. sylvestris Cyr. Open fields and woods ; rare on Eock, abundant in Spain; 11-5. I. Top of Mediterranean Steps and towards Breakneck Battery ! Above oil tanks beyond Catalan Bay ! III. ! Grows 18 in. high in the mountains ! Var. pappulosa Lge. Achenes with a short setose pappus, is said to be commoner in the province than type. I. Eavines on west slopes, K., D. III. i. South slopes of San Eoque, Boiss, D. ii. S. de Palma, Bev. iB. rotundifolia B. & E. Woods and higher mountain slopes ; locally very common ; 1-5. Willkomm and Lange and Perez Lara state that only var. hispanica Wk. is found in Spain. It is stoloniferous and has a much longer pappus. I have dug up dozens of plants and never saw a trace of a stolon, nor is the pappus long. The leaves vary from deeply reniform to truncate, and at lower elevations, where it overlaps B. sylvestris, they are 54 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD somewhat narrowed below. I suspect the two hybridise. III. i. Queen of Spain's Chair ! ii. Mountains ! Conyza amhigua DC. Waste places, roadsides and gardens ; frequent; 5-11. I.! III.! Aster longicaulis Duf. Marshes near sea ; locally rather frequent ; 9-11. III. ii. Palmones Playazo ! iii. Salt Pans ! Pulicaria arahica Cass. var. hisjxmica Boiss. Damp and waste places ; ditches and watercourses ; frequent ; 5-9. The type is not found. A dwarf form is frequent, as at Palmones Village. I. A few plants at waterworks on Willis's Road ! II. ! III. ! Var. percimans P. L. is biennial or perennant, stouter, branches shorter, subpaniculate, leaves very long, and is, I think, frequent. III. i. Railside near San Roque Station, and elsewhere ! P. dysentcrica Gaertn. f var. hispanica Wk. Marshes ; locally frequent ; 7-9. The variety is puberulous, not woolly, leaves flat, cordate, not undulate and acutely auricled. Often 3-4 ft. high. III. i. Cork Wood Sotos! ii. Algeciras, Bev. iii. Guadacorte Marshes ! Salt Pans ! P. odora Eeichb. Heaths and sandy plains, and in woods ; very common ; 5-6. III. ! Inula viscosa Ait. Eough stony ravines and watercourses ; frequent; 7-11. I. ! II.! III.! fVar. laxiflora Boiss. has heads longly peduncled and distant, in a lax panicle, and occurs frequently with the type. I. ! III. ! I. critJimoides L. Tidal rivers ; locally common ; 9-10. III. ii. and iii. Palmones River, both sides ! Aguacorte River 1 Palmones Playazo I Salt Pans ! Asteriscus viaritimus Moench. Dry rocky places ; locally abundant; 1-12, but chiefly 3-6. Varies much in size and habit. Erect specimens look very like A. aquatimis, but it is perennial, with the central head peduncled. I. ! III. i. ? Slopes of San Roque, K., D. I strongly suspect that A. aquaticus has been mistaken for it here. ii. Carnero Point ! ■\A. brachiatus Jord. & Fourr. Similar places; rare? 3-6. III. i. Algeciras, Bev. Probably at Carnero Point, whence my specimens do not differ appreciably from last, of which this may be only a slight variety. A. aquaticiis Moench. Dry fields; common; 5-6. Annual, like A. s2n7ios2is, but with more divaricate branches; soft involucre tips, and central head closely sessile. III. i. Carteian Hills ! North of San Roque ! ii. Hills round Algeciras ! Carnero Hills ! iii. Palmones Village ! A. spinosus G. & G. Dry hills and bushy places ; common; 4-6. I.! III.! Var. aureus Wk. Rays much wider. III. ii. As common as type near Algeciras, but not seen elsewhere ! Peridercea fiiscata Webb. Cultivated and fallow fields, and roadsides; very common, 12-6. Pales between florets, tube of latter flattened and winged, prolonged into a cap over achene. I. Near Inundation, K. Notthere now I think. II. ! III. ! \Antliemis arvensis L. Sandy and gravelly heaths ; occasional A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 00 or frequent ; 3-6. Pales of receptacle becoming conspicuous as flowers fade, tube of florets not prolonged, achenes ribbed, not tuberculate. Much resembles Ormenis mixta. The type and variety run into one another, but Perez Lara records var. only. I. K. II.! III. i. Bonel'sFarm! ii. Palraones Pinar ! Var. incrassata Boiss. Peduncles greatly thickened after flowering, almost tubiform. III. i. Pinar de los Bigotes! iii. Palmones Marshes ! .4. Cotula L. Similar places; occasional, or here and there abundant; 5-7. Aromatic, leaf segments capillary, pales not conspicuous, narrow, achenes tuberculate. III. i.! ii. ElCobre! A.maritima L. Sand-dunes; locally plentiful ; 5-7. Peren- nial, in dense tufts, leaf segments thick, deeply punctate. I.? K. II. K., D. Probably not also in I. III. i. Hedge at Bond's Farm, an abnormal drawn up form ! ii. and iii. Mouth of Palmones Eiver, both sides ! " Mountains above Algeciras," B. (& B. ! An unlikely station, probably the shore is meant. ■ Ormenis mixta DC. Sandy heaths, &c. ; abundant ; 5-9. Annual, prostrate and rooting, or suberect and tall, leaves narrow spathulate, terminal pinnae the longest, pales folded over achenes, tube of disc florets expanded and spurred at base. II. I III. ! Achillea Ageratum L. Dry places in rather stiff soil ; occasional; 5-10. Rigid, erect, heads yellow, small, in a compact corymb. III. i. Alcadeza ! Magazine Hill ! Malaga Gardens! Second Venta! By upper ford over Lajo ! &c. iii. Near Los Barrios Station ! Guadacorte ! Diotis maritima Sm. Sand-dunes ; locally common ; 6-7. II. 1 III. ii. Sandy Bay ! Beyond Carnero Point ! Anacyclus racliatus Lois. Fields, roadsides and waste places ; very common; 4-6. Usually tall and stout, sometimes prostrate, two outer lobes of disc florets much larger. I. ! II. ! III. ! tVar. imrpurascens DC. has ray florets red on back. I. Less common than type, K. III. i Noted, but no stations recorded ! ii. Algeciras, much rarer than type, K., D. A. clavatus Pers. Similar places; rare? 4-6. Ray white, shorter, phyllaries without a broad appendage. II. Two speci- mens, one ray less ! III. ii. Palmones Pinar! Algeciras Station ! '? iii. Hills near Los Barrios Station I [Gladanthus arahicus Cass. (C. p-olifera DC), a yellow-rayed annual, with long proHferous branches from below the central head, and leaves with linear segments, is recorded from Gibraltar by Kelaart only.] Matricaria glabra Lag. Fallow fields and waste ground ; very common ; 4-5. No pales between florets, achenes curved, with auriculate pappus, ray achenes connate with involucre, leaf segments linear, not setaceous. I. Above Willis's ! IL! IIL! [ill. discoidea DC. {M. suaveolens Buch.), a roadside casual near Algeciras bull-ring.] Prolongoa pseudanthemis Kunze. Sandy heaths ; locally common; 3-4. A small annual, leaves pectinipartite with short 56 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD entire lobes, peduncles very long, ray white, becoming carmine. III. i. By Second Pine Wood ! Chrysanthemum segetum L. Cornfields ; common ; 4-5. Leaves, at least lower, trifid, lobes deeply incised. I. Abundantly in gardens, K. III. ! Pinardia coronaria Less. Eoadsides, banks and fields ; very common ; 3-6. Eay pale yellow, or deep yellow towards base, or wholly deep yellow, the latter the rarest. I. ! II. ! III. ! ]P. anisocejjhala Cass. Sand-dunes; locally abundant; 4-6. Tall, glandular viscid, with entire or toothed leaves. III. ii. and iii. Guadacorte and Palmones Sands, both sides of Eiver! " Mountains above Algeciras," B. £ B., doubtless a misuse of collectors' printed labels. \Coleostephus Myconis B. & E. Sandy fields; rare, but perhaps mistaken for Chrysanthemum segetum ; 3-5. A glabrous yellow-rayed annual, leaves acutely serrate, achenes with a tubular crown. I. Sandy fields, K. III. ii. Near shore at Algeciras, B. d B. '''Artemisia i^ontica L. Fields ; frequent ; 7-10. Ashy green, herbaceous, 12-18 in. high, leaves short, bipinnatifid, segments linear, heads small, hemispherical, bright yellow. I have only seen very late specimens, and have not collected it. I think much more general than my notes show. [I. A doubtful native, D.] III. i. Carteian Hills, especially near Francia's Farm ! iii. About Los Barrios Station ! iHelichrysum rupestre DC. var. Boissieri Wk. {H. Fontanesii Camb.). Eocks ; locally frequent ; 5-6. Stems few or solitary, leaves broadly linear, corymbs and heads rather large, phyllaries glabrous. The description emphasises that all phyllaries are as long as florets ; they are certainly not so in our plant, and hardly difi'er from those of H. Stcechas. The plant is said to be inodorous. I have not tested this. I. All over upper Eock, especially on precipices ! H. decumhens Camb. Eocks ; rare ; 5-6. Much smaller and more decumbent, leaves close set, shorter, soon reflexed, heads smaller, fewer, deeper yellow. I. Boiss. ! -H". Stcechas DC. Eocky slopes ; locally common ; 5-7. Like H. rupestre, but strongly aromatic, much more caespitose and many-stemmed, leaves much narrower, heads smaller, in smaller clusters, often paler yellow. III. i. Cork Wood Crags! Alcadeza Crags ! H. serotinum Boiss. {H. angustifoUum DC). Sand-dunes ; locally frequent; 7-9. Densely cgespitose, many-stemmed, leaves very narrow, anthodes smaller and more cylindrical than in other species, outer phyllaries woolly on back. Leaves canescent or quite green. III. ii. and iii. About Palmones, on both sides of Eiver ! Gnaphalium luteo-album L. Sandy places ; occasional ; 3-8. I. North Front, near Catalan Bay, D. III. i. Along the Lajo ! Almoraima ! Alcadeza ! ii. ! iii. Salt Pans ! Filago germanica L. Fields and waste places ; rare ? 5-6. A FLORA OP GIBRALTAR AND THE NE1GH150URH00D 57 Erect, branched above only, branches ascending, leaves narrow, acute, not overtopping heads, heads 20-40 in cluster, obscurely angled. I. Eare, A'. iF. apiculata G. E. Sm. Similar places ; frequent ? 5-6. Upper leaves obtuse, apiculate, heads fewer in clusters, deeply 5-angled. Very near last and next, perhaps confounded with either. III. i. Campamento Common ! F. spatJmlata Presl. Similar places ; very common ; 5-6. Horizontally bi'anched from base, inflorescence proliferous, leaves broader, obtuse, undulate, heads fewer in clusters, deeply grooved. Var. prostrata Wk., prostrate, and var. erecta Wk., erect, grow together. II. D. III. ! F. arvensis L. Similar places ; rare ? 5-7. Slender, erect, leaves linear, floral shorter than clusters, phyllaries not embrac- ing achenes. II. K. III. i. Near San Eoque, Brouss. F. gallica L. Similar places ; very common ; 5-6. Floral leaves much longer than clusters, phyllaries embracing achenes. I. South and west slopes, K., D. III. i. and ii. ! Phagiialon saxatile Cass. Dry rocky places, walls and sand- dunes; frequent, abundant on Eock ; 3-6. I.! III. i. Campa- mento Common! Carteia! Cork Wood Crags, &c. ! ii. Near ElCobre! Palmones Playazo ! Carnero Point ! iii. Guadacorte! Evax aster isci flora Pers. Dry sandy soil; rare? 4-6. In- volucrating leaves acute, pales of receptacle cuspidate, anther tails short, dentate. III. ii. Algeciras, Winkl. ! The specimen looks to me more like next. E. pygmcea Pers. Similar places ; common ; 4-6. Very like last, but smaller, involucral leaves obtuse, pales acute, anther tails linear, entire. III. i. Eiver bed at Almoraima ! iii. Gardens at Salt Pans ! This or the last grows in many other places, but I did not learn to distinguish them. iE. Cavanillesii Eouy. I know nothing of this beyond Debeaux's note in Fl. p. 107. III. ii. Dry sandy hills at Alge- ciras, Bev. Senecio Lopezii Boiss. Woods ; locally frequent ; 4-5. A tall handsome species, leaves large, undivided, heads large, corymbose. iVar. minor Wk. (S. gibraltaricus Eouy), the only form recorded, seems indistinguishable from type. III. i. Cork Woods! ii. Neck above Pelayo, and slopes of El Frayle Eidge ! Summit of S. de Palma, Bev. ■''•S. Cineraria DC. is quite naturalised about Europa Glacis and Flats and by Governor's Cottage ! S. foliosus Salzm. Damp grassy places, and by watercourses ; frequent? 6-10. Near *S. Jacobcea L., but shorter, and inflores- cence laxer. Its late flowering gave me little opportunity of distinguishing it from S. erraticus. I. Catchment below Eock Gun ! A specimen or two below Breakneck Battery ! Above Main Eoad near Alameda (Jf., as S. Jacobcea). II.! Varying in leaf cutting, perhaps both species occur. III. ! fVar. snffrutescens Wk. is woody at base, leaves thicker, sub- 58 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD fleshy, peduncles stouter, bracts more numerous. I. Western slopes, Wk. S. erraticus Hevtol. Similar places ; common? 6-10. Near *S. aquaticus Huds., but with lax paniculate, not corymbose inflore- scence, leaves much less divided than in last. II. ! ? III. i. and ii. ! S. gallicus Chaix. Sand-dunes and light soil; locally common ; 1-5. A rayed annual, inflorescence corymbose, leaves all pinna- tisect. The type (var. laxiflorus DC.) has an external calycule of short phyllaries. I. At and beyond Governor's Cottage ! Var. exsquameus DC. has no calycule. III. i. and iii. Sand- dunes at Linea and Palmones ! Dautez says type and var. are indiscriminately mixed. I have found type only on the Rock, and only var. in Spain. S. petrceus B. & R. ? Rocky slopes rare ; 4-6. Annual, stout, rather tall, stem leaves entire, broadly amplexicaul, upper some- times inciso-dentate, anthodes few and large. I. ? Dasoi. This collector's records are not reliable. ]S. leucantliemifolius Poir. Sandy places ; rare; 3-4. A low annual, often much branched, leaves inciso-dentate, not lobed, corymb few-headed, heads larger than those of S. gallicus. I. Sandy grassy places on w^est side of Rock, D. S. minutus DC. f var. gibraltaricus Wk. Shady rocky places ; rare ; 5-6. Heads few, rather large, on long subradical pedun- cles, radical leaves dentate, cauline simply pinnate, lobes linear, sometimes dentate. I. Lein.l Both sides of Rock, Wk., Boiss., (&c. Kelaart indicates the Governor's Cottage as its habitat, where I have repeatedly searched in vain for it. S. lividus L. var. major G. & G. Shady bushy ground ; rather rare ; 2-6. I. Lem. ! East slopes, Wk., D. III. i. Summit of Chair ! Cork Woods ! ii. Railside near Algeciras ! Waterfall ! S. vulgaris L. Fields, roadsides and waste places ; very common; 1-12, but chiefly 12-3. I.! II.! III.! Reaches highest summits ! Calendula arvensis L. Open sandy, stony and grassy places, rarely in mountains or woods ; abundant; 11-5. Strictly annual, often subsimple and low, sometimes branched and straggling, Hke next. Outer fruits erect or incurved, lateral wings broad, spines on back many, large, in two rows, usually a spur inside at base. It varies with deep orange flowers. I. ! II. ! III. ! A field full of the orange-flowered form at Magazine Hill ! Var. malacitana P. L. Flowers half the size, ligules often not longer than phyllaries, fruit larger and more spinose. Debeaux admits the var. only, though it is much rarer than type, at least in good forms. I. Below Devil's Gap ! III. Rather frequent to Carnero Point I G. stellata Cav. Rocky and stony slopes ; locally common ; 3-5. Stout, much more straggling, probably always annual, but often becoming woody at base, flowers as large as next, outer fruits stellately spreading, muricate, not spiny on back, wingless or nearly so, no spur on face at base. I.! III. i. Alcadeza Crags! ? Perhaps only large C. arvensis, I did not see fruit. A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 59 fC S7iffruticosa Vahl. (C. marginata Willd.). Eocky and stony places near sea; locally frequent; 3-5. Always perennial, stouter and more woody, with large flowers, but little differing from last. The synonymy is somewhat involved. I. Maritime rocks near Landport, D., Bev., Nilss. O'Hara's Tower ! Var. tomentosa Ball (C. incana Willd.) is a very white silky form. I. Eocks on north-east slopes near Landport, K., D. I suspect the north-west slopes are meant, where it is plentiful, and on the debris below ! III. i. Slopes of San Eoque, D. ii. Carnero Point ! Atracty lis cancellata 1j. Dry hills; frequent; 5-6. I. Middle parts of Eock below Michael's Cave, K. Europa, Hurst ! III. i. Opposite Francia's Farm ! Carteian Hills ! Near Pinar de los Bigotes ! Path to First Pine Wood! ii. Beyond Carnero Point! Valley near Frayle Point ! Garlina gummifera Less. Grassy places ; locally common ; 8-9. III. i. From Campamentoto First Pine Wood ! ii. Carnero Hills ! fC. lanata L. Grassy and stony hills ; rare; 6-8. Like next, but stems simple or branched at top, leaves broader, heads solitary, 1-lJ in. in diameter over ray, inner phyllaries purple. III. i. South slopes of San Eoque, D. G. racemosa L. (C. sulphur ea Desf.). Similar places ; abundant ; 7-9. II.! III.! C. corymhosa L. Bushy and heathy places, chiefly on moun- tains ; frequent ; 6-7. Much more leafy, with larger heads, 1 in. broad. I.! III.! Var. involucrata Boiss. (var. major Lge.) is stouter, sparingly branched at apex only, heads l|-2 in., floral leaves longer than ray. I think frequent, but have not collected it. I. Herb. Madrid, teste Deb. Kentrophyllum lanatiim DC. Open fields ; very common ; 6-8. Much resembling Carlina racemosa when not in flower. Arachnoid-pubescent, leaves glandular-viscid, outer floral bracts erect patent, as long as flowers, heads 1 in. long, achenes and pappus pale. III. i. and ii. ! K. bcBticum B. & E. Similar places ; rare ? 6-9. Stem white, leaves shining, subglabrous, the floral recurved patent, longer than heads, achenes and pappus blackish. Debeaux admits this species only, though it appears much the rarer. I. Nilss. III. i. Eough slopes of San Eoque and S. Carbonera, D. Lane beyond Bonel's Farm ! iK. arborescens Hook. Bushy slopes ; locally common ; 5-6. I. Chiefly south-west, less common north-west, still rarer east slopes ! III. i. Carteia, K. Carduncellus co&ruleus DC. Eough hills and fields; com- mon ; 4-6. Var. mcisus DC. has all leaves pinnatifid and spinose dentate, and is said to be commoner. My records are for the aggregate. I. Near Devil's Tower ! III. ! Chiefly in i. ! Onopordon Acanthiiim L. Waste places and roadsides ; rare ; 60 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 7-8. Perez Lara only gives one other station in the province, but says that 0. nervosum Boiss., with glandular, not glabrous corolla is much more frequent. III. i. Between Neutral Ground and Guadarranque, D. Campamento and San Eoque, Frere. Bourgcea humilis Goss. Dry grassy places ; very common ; 5-7. Often mistaken for next. Inner phyllaries longly acuminate and spinose, like outer. An albino is rather frequent, and is doubtless the plant referred to as Cynara alba by Kelaart. I. Engineer Eoad ! III. ! Cynara Cardunculus L. (C. horrida Sibth.). Similar places ; rare or error ? 5-7. Much taller, with fewer broader leaf segments and rachis, phyllaries with very stout spines, the inmost with an expanded membranous appendage. I much doubt its occurrence. [I. Boiss. ex Deb. Fl. p. 113. Boissier's label reads "between Estepona and Gibraltar"! Kelaart's specimen is Bourgaa humilis.'] III. i. Dry hills of Campamento, K. 7 San Eoque, D. ? Notobasis syriaca Cass. Fields ; common ; 5-6. II. A plant or two ! III. ! [Cirsium lanceolatum Scop. var. hypoleucum DC. is recorded by Kelaart as introduced in Gibraltar. I have seen leaves of what might be this between Guadacorte and the railway bridge.] C. giganteum Spr. Eather damp places, chiefly near streams; here and there plentiful ; 6-8. Often 8-9 ft. high. III. i. First Pine Wood ! Eare in Carteian Hills ! Near Second Venta ! Between Malaga Gardens and Alcadeza ! Near Almendral, K. ii. Below El Cobre ! Near top of Waterfall Valley ! M. de la Torre ! Carnero Hills ! \Carduus myriacanthus Salzm. Sands near sea; rare; 4-5. About 1 ft. high, like a very spiny C. ijycnocephalus, but phyllaries much narrower, with long spinose tips. G. Beuterianus Boiss., which differs in no important character, is frequent in the province. II. Isthmus of Gibraltar, Wk. ! III. i. Linea, Porta (& Bigo ! About San Eoque, Boiss. C. tenuiflorus Curt. Waste places and roadsides ; frequent or common ; 4-6. Perhaps only a variety of next, with stem winged to top, heads subsessile, densely capitate, &c. I. ! III. ! G. ivjcnocephalus Jacq. Similar places ; frequent but less so than last ; 4-6. Stem interruptedly winged, naked at top, heads larger and laxer, 2-3 together, the central usually peduncled. I. ! II. ! III. ! fC. nigrescens Vill. Heathy places; rare; 4-7. Like next, but heads smaller and erect, phyllaries more glabrous, not reflexed. I. ? Dasoi. III. i. Path to First Pine Wood ! fC nutans L. Similar places; rare; 5-7. III. ii. Near Algeciras, Clem. Silybum Marianum Gaertn. Eoadsides and waste places ; rather frequent ; here and there abundant ; 4-6. I. Europa Flats ! Jews' Cemetery ! North Front ! II. ! III. ! Galactites tomentosa Moench. Dry rough fields ; abundant. A FLORA OP GIHRALTAR AND THE NEIGHBOURHOOD 61 often in large masses ; 4- 6. I. Lower and middle parts, rarely upper Eock! II.! III.! fVar. integrifolia Boiss. Leaves entire or denticulate. It may be common, but I have not seen it. I. K. Serratula hcetica Hoiss. yur.pinnatifoUaV^k. Wooded slopes; locally frequent ; 6-7. Like a dwarf Centaurea, but phyllaries long, lanceolate, acuminate into a spine. III. i. Queen of Spain's Chair ! ii. Slopes beyond Waterfall ! Carnero Hills ! Leuz&a conifera DC. Wooded slopes and heaths ; rare ; 5-6. III. i. Wooded slopes of San Roque, I). Alcadeza Plain ! Grupina vulgaris Cass. Stony slopes; rare; 5-6. III. i. Alcadeza Crags ! Microlonchus Clusii Spach. Roadsides and bushy banks ; frequent ; 4-7. I. Jews' Cemetery ! Rosia Parade ! III. ! Grows 5 ft. high by the Miel ! Centaurea tagana Brot. Woods ; locally occasional ; 5-6. One of the largest of the genus, with very large heads of dingy yellow flowers. III. i. First Pine Wood ! Cork Woods ! ii. Upper slopes of Waterfall Valley ! fC. alba L. var. deusta DC. Dry rocky or sandy slopes; locally frequent ; 6-7. Habit of C. aspera, but phyllaries lax, ovate, broadly hyaline, with a blackish brown central band. III. i. Alcadeza Crags ! Cork Wood Crags ! \G. sempervirens L. Woods; rare; 6-7. Tall, much branched, leafy to apex, flowers purple, appendages shortly pectinate. III. ii. Waterfall! \C. uliginosa Brot. Marshes; rare; 6-7. Subsimple, 4-6 ft., leaves very long, narrow, peduncles long, pectinae shorter and fewer than in last, and erect or spreading, not reflexed. My specimens are taller than the description gives, and have the upper leaves quite entire, not denticulate. III. i. Almoraima Soto ! C. imllata L. Grassy and bushy places, sides of ditches, &c. ; very common ; 2-5. Dwarf, very leafy, flowers pink or rosy, phyllaries pale green with black edges, appendages reflexed, pectinate. Varies much in habit and leaf lobing. I. ! II. ! III. ! C. cUluta Ait. Rough fields ; rare ; 5-7. Phyllaries pale, appendages white, shortly lacerate, subdecurrent, terminal 1-3 spines rigid, not longer than others, and erect. III. i. Carteian Hills, a single plant I fC. Seridis L. Sandy ground; rare; 4-6. Perennial, erect, branched, woolly-canescent, leaves longly decurrent, the lower dentate, rarely lobed, heads large, florets purple, phyllaries with 7-11 remarkably long reflexed spines. III. i. Near San Roque, Ball. t Var. maritima Lge. differs in very large lyrate-pinnatifid leaves, the cauline less decurrent. III. ii. Sea sand at Algeciras, P. L. C. sonchifolia L. Sand-dunes ; rare ; 5-6. Somewhat like C. sphcerocepJiala, but scabrid-pubescent, leaves shortly decurrent, heads large, disc florets white, phyllaries with 5 rather short reflexed spines. II. Sea sand at Gibraltar, Brouss. Probably on Neutral Ground. C. aspera L. Dry stony and sauidy hills ; locally common ; 62 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 5-7. Much branched, 1-2 ft. high, leaves more or less asperous and cinereous, narrow, usually pinnatiiid with narrow lobes, phyllaries pale green, spines 3-5, palmate. III. i. Cork Wood Crags ! Alcadeza Crags ! C. splicBrocephala L. Sand-dunes ; very common ; here and there abundant ; 5-6. Prostrate, heads scarcely radiant, spines of phyllaries 5-7. I. Beyond Catalan Bay! Sentry Fence! II.! III.! C. polyacantha Willd. Woods, hedges and heathy places ; rather frequent ; 4-6. Leaves more lobed than last, rays much longer and brighter, spines of phyllaries 9-13. I. Findlay ! ? I believe this species, but not labelled. II. K., D. III. i. Carteia, K. Second Pine Wood ! Pine Wood Plains ! Near San Eoque Station! Cork Woods ! Beyond Alcadeza! ii. Palmones Pinar and Playazo ! iii. Guadacorte ! Palmones, Bev. [C. acaulis Haens. (C. Hcenseleri B. & E.) is reported by Kelaart from the Carteian Hills.] C. Calcitrapa L. Waste places and roadsides ; very common ; 6-8. I. Europa and Windmill Hill Flats! North Front! Glacis ! II. ! III. ! C. melitensis L. Eoadsides and cultivated fields ; locally frequent ; 5-6. Stem winged, florets yellow, glandular. I. From Middle Gate to Jews' Cemetery ! Between Ince's and Castle ! III. i. Near Malaga Gardens ! Path to First Pine Wood ! Campamento Common ! ii. Dry hills at Algeciras, Bev. [C. solstitialis L. a similar species, but spines much longer and florets eglandular, was found by Kelaart on the Glacis, introduced.] EcJmiops strigosus L. Cornfields ; locally frequent ; 5-6. III. i. Around San Eoque, especially on north ! ii. Near Algeciras, Bev. Scolynms maculatus L. Fields and roadsides ; very common ; 5-6. I. Cultivated fields on west slopes, K., D.; not there now, I think ; there are only one or two small cultivated fields left. By Haynes's Foundry ! II. ! III. ! S. hispanicus L. Similar places; very common; 5-7. I.! II.! III.! Cichorium Intybus L. Fields and waste places; common; 5-6. Varies greatly in stature and habit. Only var. clivaricatum DC. is admitted by Debeaux, which Perez Lara observes is in- separable by any constant character, and the type certainly occurs. Dwarf prostrate forms are frequent, usually the result of browsing by cattle. I. Europa ! South Barracks ! Queen's Eoad ! North Front ! II. ! III. i. and ii. ! Tolpis barbata Gaertn. Fields, woods and mountains ; very common ; 4-6. III. ! Chiefly in i. ! Var. grandiflora Ball is dwarf, leaves twice as broad, inciso- dentate, heads large, central florets purplish. III. ii. Sands near Algeciras, Ball. Hedypnois pendula DC. Grassy and bushy places ; very frequent, at least locally; 3-5. The first three species of this genus are in much confusion. H. pendula and H. tubaformis, as A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 63 extremes, are very distinct, but are connected by many inter- mediates, varying in habit, hispidity, leaf cutting, and inflation of peduncles. I only cite stations in which I have collected specimens. H. pendula is erect, usually glabrous, very slender, buds nodding, leaves bright green, flaccid. I. Common on lower parts ! III. i. Puente Mayorga ! YiiY.pmnatifida DC. has leaves pinnatifid and phyllaries hispid at tips. Probably frequent. I. Eeclamation Road ! Var. rhagadioloides Lge. (sub H. polymorpha) is similar, but hispid all over. I. Europa Flats ! I think frequent elsewhere. H. cretica Will. {H. jjolymorpha DC.'?). Open grassy and sandy places ; very common ; 3-5. Always more or less pros- trate, peduncles somewhat inflated, but not more than one- third the diameter of fruit heads. Varies as much in leaf cutting and hispidity as last, but leaves never bright green, usually thick. I. Europa Flats ! Above Alameda ! North Front ! Sentry Fence ! II. ! III. i. Near Linea Cemetery ! (with pinnatifid leaves). Near Pinar de los Bigotes ! H. tubceformis Ten. {H. cretica var. subacaulis DC). Similar places ; very common ; 3-5. Perhaps a variety of last, with peduncles greatly inflated, at least half diameter of fruiting heads. III. ii. Path near Reina Cristina Hotel, and elsewhere ! H. arenaria DC. In deep sand ; frequent ; 4-6. Very like a HypochcBvis ; anthodes much larger than in last three, and pappus rays always numerous. I. West side and sea sands near Old Mole, K. ; not there now, I think. East side, Boiss. Sentry Fence ! II. ! III. i. Linea ! Carteian Hills ! ii. Palmones Playazo ! Sandy Bay ! iii. Palmones Sands ! [H. pygmaa Wk. I gathered a few specimens of a very small species near the road and Spanish Lines on eastern side of Neutral Ground, which may belong to this species, but they were in flower only, and I was unable to find them again to get fruit.] \Hyoseris scabra L. {H. microcephala Cass.). Dry rocky and gravelly places ; rather rare ; 3-5. I. Buena Vista to Europa Flats ! III. i. Slopes of San Roque, D. H. radiata L. Similar places ; abundant on Rock, rather common in Spain. I. ! III. i. About San Roque ! Queen of Spain's Chair ! Cork Woods ! ii. and iii. ! \Bhagadiolus stellakis DC. Grassy fields or bushy places in light soil; occasional; 4-5. The type has lower leaves oblong- lanceolate, dentate, not lobed. I. K. III. i. Foot of San Roque, K., D. Carteian Hills ! Alcadeza ! Almoraima ! ii. Algeciras Station ! M. de la Torre ! Var. edulis DC. has leaves lyrate pinnatifid, the terminal lobe large, orbicular. I think as common as type and mixed with it ! Thrincia hispida Roth. Grassy and rough places ; both varieties abundant ; 1-12, chiefly 4-5. Var. minor Boiss. is smaller in all parts, scapes 1-3 in. I. ! II. ! III. ! Var. viajor Boiss. is more hirsute, scapes 3-12 in., thickened at apex, heads twice the size. I. ! II. ! III. ! 64 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD T. tuberosa DC. Open grassy fields ; common ; 10-4. All achenes with beak about half their length. III. i. and ii. ! T. maroccana Pers. Similar places ; rare ? 5. Annual, outer achenes shortly, inner very longly beaked, but this is variable. It may be only a first year's flowering of last. I. Frere. III. i. Near San Eoque, Boiss. jLeontodon liispanicus Mer. var. psilocalyx Wk. Very hispid, all achenes with pappus, not inner only. III. ii. Sandy hills near Algeciras, Winkl. Helminthia echioides Gaertn. Hedges and bushy places ; occasional; 5-9. Outer phyllaries cordate, as long as lanceolate inner. I. Scud Hill, Dumbreck. Gibraltar, K. III. i. Abun- dantly by San Eoque Eoad, K. ii. Algeciras, Rev. H. comosa Boiss. Similar places ; rather common ; 5-6. Stem and leaves with much larger and more numerous asperities on tubercles, outer phyllaries ovate or lanceolate, one-third as long as inner. II. ! III. ! A dwarf form is abundant on Carnero Hills! Urospermum picroides Desf. Dry bushy, grassy and sandy places ; common ; 3-5. 1. 1 II. ! III. ! Picris hieracioides L. Stony and rocky places ; rare ; 6-7. III. i. Eoadsides near San Eoque, D. Podospermxim calcitrapifolium DC. Grassy places and by ditches; occasional or frequent; 4-5. III. i. Carteian Hills ! San Eoque ! Foot of Chair ! iii. Salt Pans ! Near Palmones Village, often in a very dwarf form ! Scorzonera liispanica L. var, latifolia Koch. Cornfields ; rare ? 4-6. Phyllaries very unequal, leaves broad. III. iii. Between San Eoque and Algeciras, Winkl. fVar. glastifolia Wallr. has leaves about 2 lines broad, and is the only form I have seen. It is frequent. III. i. and ii. ! Tragopogon porrifolms L. ? Sandy and grassy places ; rare or next mistaken for it? 4-5. Flowers purple, achenes abruptly narrowed into a long beak, pappus of all achenes plumose. I. Brouss. An escape ? tVar. australis P. L. Leaves undulate, florets much shorter than phyllaries, achenes gradually attenuate into a beak shorter than fruit. III. i. Near San Eoque, D., who does not record next; a possible error ? Geropogon glaber L. Cornfields and grassy places ; locally frequent ; 4-5. Flowers pale purple, pappus of outer achenes of 5 short pales, not of hairs. III. i. Hills over San Eoque Station ! ii. Eailside near Algeciras! Towards Sandy Bay! Hypockaris radicata L. Sandy places ; frequent or common ; 4-6. Tall, often 2 ft. high, scapes elongate, erectly' branched, all achenes with beak longer than fruit. III. ! Var. heterocarpa Mor. has outer achenes not beaked. III. ii. About Algeciras, Bev. H. Salzynanniana Coss. Sand-dunes ; very common ; 2-6. Annual, flowers as large as last, phyllaries hispid or glabrous, leaves usually hispid-ciliate. Sometimes regarded as a variety of H. glabra, but very distinct. II. ! III. ! A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 65 in. glabra L. Sandy fields ; frequent? 2-6. Annual, glabrous, flowers small, closing at noon. Varies in length of beak of outer and inner achenes, but no varieties recorded in our limits. III. i. and ii. ! [Seriola atnensis L. A specimen from Willkomm at Kew labelled " Malaga and Gibraltar" was probably gathered near the former place. Kelaart records it from the neighbourhood of Gibraltar.] Taraxacum officinale Wigg. {T. Dens-leonis Desf.). Grassy places and roadsides ; rare ; 5-6. Kelaart records var. Icevigatimi DC, with runcinate pinnatifid leaves, and var. ohovatum DC, with them entire, from the Eock. There is little doubt that he mistook Hyoseris radiata for it. [I. K.] III. ii. Algeciras, Bev. Lactuca tenerrima Pourr. Eocks and old walls ; locally very common; 6-7. Usually as fvar. scabra Boiss., with scabrid white hairs, but the type occurs. I. ! III. ii. Walls at Algeciras and El Cobre ! L. saligna L. Ditches and banks ; rare ; 7-9. Stem leaves narrow, entire, sagittate, the radical often sinuate-lobed. III. i. Eailway about San Eoque Station ! ii. Eailside near Algeciras ! ? iii. Salt Pans ! ? Guadacorte Marsh ! ? Leaves only seen in last three stations, which may have belonged to L. Scariola L., a common species in the province. Picriclium tingitanuyn Desf. Sands by sea, and rocky places ; frequent ? 4-6. Usually short, phyliaries broadly ovate, squarrose, the outer acuminate, conspicuously broadly scarious, white-edged. I. Clefts of rocks on west slopes, K., D., Bev. II. ib. III. ii. Algeciras, Bev. P. intermedium Schultz. Sandy and gravelly places ; com- mon ; 2-5. Annual, slender, leaves thin, green. I. ! III. ! P. vulgare Desf. var. crassifolium Wk. Eocks and sandy places near sea ; rather frequent ; 1-12, chiefly 3-5. Perennial, phyliaries lanceolate, outer not squarrose, nor broadly winged. The type, which is not recorded, has the habit of P. intermedium, the var. is dense and compact, with thick fleshy leaves. I. ! II. ! III. i. San Eoque, Golm. ii. Palmones Playazo ! Sandy Bay ! Carnero Point ! fVar. maritimum Boiss. Has all leaves pinnatipartite, with narrow segments. I. East and south slopes, Boiss., K., D. III. ii. Palmones Playazo ! Sonchus tenerrimus L. Eocks, bushy places, and old walls ; very common ; 1-12, chiefly 2-6. Varies greatly in leaf cutting and size of flower. tVar. IcBvigatus Lge. (var. anmms Lge. ?). Annual, peduncles gla- brous, leaf segments narrow, acute. I. D. III. ii. Waterfall Valley ! Var. glandulosus Lge. Annual, peduncles densely glandular, leaf segments broadly oval, obtuse. I. Wk. III. ii. Near Algeciras, Wijikl. fVar. spinulosus Lge. As last, but leaf segments acute or acuminate, much spinulose dentate. III. i. Alcadeza Crags ! Var. perennis Lge. Perennial, peduncles and heads glabrous Journal op Botany, May, 1914. [Supplement.] / 66 A FLORA OP GIBRALTAU AND THE NEIGHBOURHOOD or glandular, leaves various. Much the commonest form. I. ! III. i. Carteia ! Alcadeza ! Malaga Gardens ! ii. from Pal- mones Playazo to Carnero Point ! S. oleraceus L. Eoadsides, grassy places and by streams ; frequent ; 1-12. Leaf auricles acute, defiexed or spreading, achenes rugose. I, ! II. ! III. i. and ii. ! S. as])er Vill. Similar places ; less frequent ; 3-9. Auricles defiexed and rounded, achenes ribbed, not rugose. I. Eare, K. II. ! III. ! S. glaucescens Jord. Sandy cultivated fields ; rare ; 3-5. Biennial, leaves thick, heads tv^ice the size of last, achenes more or less winged, retrorsely ciliate. III. ii. Near Algeciras, Wmhl. j^theorrhiza bulbosa Cass. In deep sand, occasionally stony places ; common ; 3-5. I. Near O'Hara's ! Near Willis's ! Catalan Bay ! II. ! III. ! \Barkha2tsia fontida DC. Old walls and waste places; rare? 6. Like next, but outer achenes more shortly beaked, flowers sub- solitary, nodding in bud, peduncles thickened at top. I. South and west slopes, K., D. B. taraxacifolia Thuill. Fields and rough bushy places ; abun- dant ; 12-5. Most variable, either csespitose and spreading, or erect, with solitary stems branched at top. I. Chiefly lovv^er north and west slopes ! III. ! fVar. Haenseleri Boiss. Glabrescent, leaves obtuse, dentate, not lobed. I. ! Eare ? Crepis tingitana Ball. Woods and heathy slopes; locally fre- quent ; 3-5. Like a Hieracium, little branched, leaves few, large, basal subtruncate, cauline narrowed below. III. i. Cork Woods ! Majarambout Crags ! ii. Mountains ! C. virens L. Dry sandy places, occasional ; 3-5. Unlike any British form, but agreeing closely with the eastern C. parviflora Desf. Erect, corymbosely branched, leaves mostly radical, dentate, flowers rather small, pale yellow, often reddish on back, fading to orange. I. Above the Library ! North Front ! II. ! III. ! Chiefly about Palmones and Guadacorte ! Var. runcinata Bischff. has basal leaves runcinate pinnatifid. III. i. Grassy fields at San Eoque, D. \C. corymhosa Ten. Similar places? rare; 6. Annual, more or less pubescent, 1 ft. or more high, leaves spathulate, runcinate. Anthodes 4 lines long, on long slender peduncles in lax paniculate cymes, achenes not beaked. III. ii. S. de Palma. Bev. fVar. hcetica Wk. has stems more pubescent, leaves sinuate dentate, heads smaller, and phyllaries glabrous inside. III. ii. With type, Bev. Andryala integrifolia L. Dry rocky or bushy places ; common ; 5-7. Biennial, canescent and softly tomentose, receptacle with very long setae, pappus as long as phyllaries. The type (var. corijm- bosa Wk., A. parviflora var. latifolia Boiss.) is much branched at top, heads 4 lines, in a compact corymb, leaves entire or only dentate, often undulate. I. ! II. ! III. i. and ii. ! Var. sinuata Wk. Leaves nai'row, more or less sinuate- A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 07 dentate, or runcinate pinnatifid, cymes laxer, heads smaller. I. Common, K., D., Boiss. III. ii. Algeciras, Bev. A. arenaria B. & R. Sandy places ; frequent, locally common ; 3-5. Annual, much more softly tomentose, receptacle with short setuli, pappus shorter than phyllaries, corymh small, dense, flowers orange, red on back. II. K., Boiss. III. i. Cork Woods ! Pine Wood Plains ! Foot of San Roque, K., Boiss. A. laxiflora DC. Rocky slopes; rare; 6. Like last but anthodes 5 lines, on longish peduncles in a lax corymb, flowers pale yellow. I. D. Xanthium macrocarpum DC. Sandy, rather damp places ; locally abundant ; 7-9. Like X. italicum but with much stouter fruit spines, apical beak incurved. II.! III. i. Punta Mala! ii. Palmones Playazo ! X. sjnnosum L. Roadsides and sandy waste places ; occasional ; 4-9. II. ! III. i. Campamento and roadside to San Roque ! Lajo banks ! Near San Roque Station ! ii. About Algeciras ! iii. Palmones Village ! X. italicum Moretti. Rather damp sandy places and by rivers ; rather rare ? 7-9. II. ! ? Only detached fruit seen. III. ii. Pal- mones Playazo ! Campanulace^. Laurentia Michelii DC. Stream beds and under damp rocks ; rather frequent; 4-6. III. i. Queen of Spain's Chair, chiefly east slopes ! ii. Mountains ! Lobelia urens L. Streams and damp hollows ; frequent in mountains, occasional elsewhere ; 4-6. Var. longibracteata P. L. has bracts longer than calyx, some- times than corolla, and longer calyx lobes. It is said to be the common form, but the type certainly occurs frequently, and I have not distinguished them. III. i. Hedges at Campamento in profusion, K. Queen of Spain's Chair ! Alcadeza ! Cork Woods ! ii. Mountains ! iii. Palmones Sands ! \Jasione montana L. Sandy and heathy places ; frequent ; 4-6. Varies greatly in habit, duration, and length of calyx segments and pedicels. The type is not recorded. Var. echinata Wk. (var. dcntata DC). Biennial, stout, erect, 12-18 in. The usual mountain and wood form. I. Boiss. III. i. Queen of Spain's Chair! San Roque, B. (& B. Cork Woods ! ii. S. de Palma, Bev. El Cobre ! Var. bracteosa Wk. (var. littoralis Boiss., /. blepharodon B. & R.). Shorter, more caespitose, with different calyx segments. The form of sandy open spots. I. Catalan Bay ! Mediterranean Steps, K. III. i. San Roque, Boiss, D. My 1306, a very dwarf caespitose hispid form, from near La Tunares, probably belongs here. i J. rosidaris B. & R. Heathy hills; rare; 4-6. Perennial, basal leaves rosulate, heads large, calyx segments lanceolate, rigid, longly pungent acuminate, thrice as long as tube. The station cited is the classic one, but I have referred all I have seen there to J. mon- tana var. echinata. III. i. Queen of Spain's Chair, B. it B., D. 68 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD Campamda mollis 1j. Eocks ; locally frequent ; 5-7. I.! fVar. microphylla DC. has smaller ovate acute subdentate cauline leaves, and is mixed with the type. I. Broiiss. C. dicliotoma L. fvar. brachiata A. DC. Dry sandy baaks; rare ; 5-6. Annual, erect, calycine appendages narrow, as long as tube, but they vary in C. mollis, and dried specimens are not always easily distinguished. I. Gouan ! and by an unknown collector, both labelled C. mollis, but placed in this cover at Kew. III. i. Escarpments of San Eoque, D. C. Erinus L. Eocks, old walls and dry stony places ; common ; 4-5. I.! III. i. Cachon ! First Venta! ii. On railway! Water- fall ! El Cobre ! iii. Palmones ! C. Bapunculus L. Bushy places in woods and hills ; com- mon ; 4-6. Varies in habit, the strict forms usually more hispid than the diffusely branched ones, but the varieties are not clearly distinguished. I. Near Bruce's Farm ! III. ! C. jpatula L. ? Woods and bushy places; rare or error? 5-7. Biennial, stems rather stout, strict, basal leaves rosulate, corolla reddish violet, lobes rather long and narrow, widely spreading. I. Brouss. ! a poor specimen, which I think belongs to the next species. C. Loeflingii Brot. (C. erinoides L.). Sandy bushy places, occasionally in mountains ; common, often abundant ; 4-6. Annual, slender, much branched from base, without a rosette, corolla smaller, less open, black at the base, not white as described. Its habit and appearance are very different from C. patula. I. Not very common, chiefly on higher parts, K. III. ! My 2032, in fields towards S. Lorca, is shorter and more csespitose, the corolla without a dark base, and calyx segments much broader. I suspect it belongs here. f Var. filiformis Lge. Very slender, elongate, retrorsely scabrid, branches elongate, calyx segments scabrid, very longly setaceous. III. ii. Sands near Algeciras, Kusinshy. Trachelium caruleimi L. Eather damp walls and banks ; occasional ; 5-6. III. i. Lajo near upper ford, and above First Pine Wood ! Eailside beyond Almoraima ! ii. By E. Lobo ! Roadside at Puente de los Pastores ! Waterfall Valley ! Ericace^. Arbutus Unedo L. Woods and mountain slopes ; locally fre- quent; 10-12. III. i. Cork Woods I ii. Mountains to highest ridge! Bhododendron bcetimim B. & E. Mountain valleys ; locally common ; 4-6. III. [i. Castellar, beyond our limits, K.] ii. Mountains to highest ridge ! Erica ciliaris L. Heathy places in mountains and woods; occasional; 7-10. III. i. Queen of Spain's Chair! ii. Mountains! E. arborea L. Similar places ; much commoner ; 3-4. III. i. Cork Woods 1 ii. Mountains ! E. scoparia L. Similar places ; very common ; 4-5. III. i. Queen of Spain's Chair 1 Cork Woods ! Alcadeza Crags I ii. Mountains ! A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 60 E. austraMs Ij. Similar places; common; 12-1. III. i. and ii. Same stations, also Palmones Pinar ! E. umhellata L. Similar places ; occasional ; 2-6. III. i. Cork Woods, especially near Second Pine Wood ! Alcadeza Crags and Plains ! ii. Mountains ! Var. suhcampanulata DC. Throat of corolla more open, stamens shorter. III. ii. S. de Saladillo, P. L. fVar. viajor Coss. Corolla nearly twice the size, 2|-3 lines long, anthers larger but less exserted. Ill, i. Near San Eoque, Wk. ii. S. de Palma, Bev. fVar. ananclra Lge. Corolla larger than type, cylindrical, anthers none. III. ii. S. de Luna near Los Barrios, Nilss. E. viediterranea Lt. Similar places ; rare; 1-3. III. ii. Hills behind Algeciras, Frere. Calluna vulgaris Salisb. Similar places ; abundant ; 7-13. III. i. and ii. ! fVar. depressa W.-Dod. in Journ. Bot., 1914, p. 13. Prostrate, stems tortuous, intricate. III. i. All over Bonel's Farm ! Oleace^. Jasminum fruticans L. Eocky bushy places ; very common ; 3-5. I. ! Olea europcea L. Mountain and hill slopes ; common ; 5-6. [I. A few trees cultivated in Kelaart's time.] Var. oleaster DC. The wild form, with much smaller fruit. I.! III.! Phillyrea latifolia L. Woods ; occasional, or locally frequent ; 2-4. Described as a tree, 18-24 ft. high ; I have only seen it as a bush, 8-10 ft., often much less; leaves rather deeply and very acutely serrate, or almost entire on the same bush, fruit globose, umbilicate at apex. III. i. Cork Woods near Second Venta ! ii. Waterfall Valley ! f Var. ohliqua Ait., with elliptical lanceolate leaves, teeth almost obsolete. III. ii. With type in S. de Palma, Wk. P. media L. Similar places; rare? 12-1. Very like last, but a rounder fuller bush, with finely toothed leaves, about twice as long as broad, fruit globose, abruptly mucronate, not um- bilicate. I. Old Man's Garden ! Near Corsican's Post ! Levant ! I think not common all over the Eock, as Kelaart and Dautez report. III. ii. Eecorded in my notes, but authority mislaid. P. angustifolia L. Similar places ; common ; 1-4. Leaves almost linear. III. i. Malaga Gardens ! Alcadeza Crags ! Cork Woods ! ii. Palmones Playazo, a broad-leaved form ! Mountains ! iii. Palmones Sands ! Fraxinus angustifolia Vahl. Woods and hedges ; rather frequent ; 1-2. [I. Engineer Eoad and about Alameda, not native '?] III. i- Alcadeza ! Eailway near Guadarranque Eiver ! Cork Wood Sotos ! ii. M. de la Torre ! Near Palmones railway bridge ! iii. Guadacorte ! 70 a flora of gibraltar and the neighbourhood Apocynace^. Vinca media Hoffm. & Link. Banks and bushy places ; abundant ; 12-4. I. ! III. ! With white flowers near Levant, below Signal Station, and railside near Long Stables ! Nerium Oleander L. By streams and rivers ; very common ; 5-9. [I. Native, K.] I think this doubtful. III. ! ASCLEPIADACE.S;. ■•'\Go')np}iocarpus fruticosus E.Br. Among bushes; rare; 5-6. III. ii. At El Cobre, quite naturalized ! Gentianace^. Chlora perfoliata L. Hill slopes and banks ; occasional ; 5-8. Cauline leaves connate for their whole width, calyx divided to base, segments linear-subulate, shorter than corolla. I. Near Signal Station Eoad! Catalan Bay, Ifwrs^ ! III. i. Upper Lajo and Alcadeza I ii. El Cobre, both very slender forms ! Carnero Hills ! iii. Palmones ! Var. sessilifolia Griseb. More slender, shorter, leaves longer, more acute, scarcely connate except uppermost. Connects with next through its var. lanceolata Koch. III. i. Woody places about San Koque, D. ii. Algeciras, Bev. C. imperfoliata L. f. Similar places, partial to damp ; occa- sional; 5-6. Lower and middle leaves not connate, flowers longly peduncled, calyx segments broadly linear-lanceolate, connate at base, longer than corolla. II. ! III. i. Bond's Farm ! iii. Palmones Sands ! Guadacorte ! Gicendia filiformis Delarb. Eather damp sandy places ; locally frequent ; 4-6. III. I. Bond's Farm and lower east slopes of Chair ! ii. Near Palmones Pinar ! Eare in mountains 1 fC pusilla Griseb. Similar places; rare; 5-7. III. i. Near San Eoque, Ball ! ii. S. de Palma in Los Barrios district, Bev. Erythrcea maritima Pers. Dry grassy and sandy places ; rather frequent ; 4-6. Flowers yellow. III. i. and ii. ! Not seen north of San Eoque, but occurs in Algeciras mountains ! E. spicata Pers. t var. glauca Eev. Salt marshes ; locally frequent ; 6-9. III. iii. Salt Pans ! Guadacorte marshes ! E. ramosissima Pers. [E. pidchclla Hornem.). Sandy banks and grassy places, partial to damp ; locally frequent ; 4-8. Varies much in habit. Starved forms from Guadarranque marshes may be f. gracilis Daut. & Deb. III. i. Between Neutral Ground and Guadarranque Eiver, D. ii. Hills round Algeciras ! Carnero Hills ! iii. Palmones Sands and Marshes ! E. latifolia Sm. var. tenuiflora Hoffm. &Link. Similar places; rare ; 4-8. Eesembles last, but much taller, much branched above only, with subcorymbose inflorescence, leaves broader. Ill', ii. Saline marshes and sands at Algeciras, Bev. iii. Guadacorte Marshes ! \E. Barrelieri L. Dry or rocky hills ; rare ; 6-8. Distinct from all in its long linear-subulate leaves, and large pedicelled flowers. III. i. S. Carbonera, D. A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 71 E. Centaurium Pers. Dry rough and bushy places, and woods; occasional or frequent; 5-7. Flowers sessile, medium- sized, always in a flat-topped more or less dense corymb. I. A large- flowered form from east slopes below Middle Hill may be the true var. grandiflora Pers. (non Biv.), and a dwarf form with a rosette of very broad leaves, from the foot of the precipice below, may be the same ! Type not seen. III. i. and ii. ! A white or very pale pink-flowered form replaces the normal about El Cobre ! Var. suffruticosa Griseb. has stem subwoody at base, and shorter corolla lobes, but most specimens I have seen so labelled are quite herbaceous at the base. I. Not frequent, K. III. i. In great abundance on road to Cork Woods in company with E. major, K. No Erythrcsa is abundant there now, but type E. Centauriwn is fairly frequent. San Eoque, Boiss. iii. Hills near Los Barrios, P. L. Perhaps outside our limits. E. grandiflora Biv. {E. Boissieri Wk., E. sanguinea Mab. ?, E. Centaurium var. grandiflora Pers.?). Similar places and by ditches; frequent in Spain, rather rare on Eock; 5-7. Flowers very large, in a lax cyme, not a compact corymb. I. ! III. ! \E. acutiflora Schott. Similar places ; rare; 6. Short, leaves very acute, corolla lobes narrow, acute. The description hardly differentiates it from E. tenuiflora. III. i. By streams of San Koque, Schott. ii. Damp places at Algeciras, Bev. [BlGNONIACE^.] [Gatalpa sijringcBflora Sims is only cultivated.] Con VOL VUL ACE.E . Convolvulus althceoides L. Dry fields, &c., abundant ; 4-6. I! II.! III.! C. arvensis L. Fields and sandy places; very common; 5-8. I.! III.! f Var. linearifolius Choisy, with elongate broadly linear leaves, is alone recorded by Debeaux, but the type is much more fre- quent. I. K. f C. siculus L. Dry bushy and stony places ; frequent on the Eock ; 3-5. I. ! III. i. Queen of Spain's Chair ; rare ! tC undulatus Cav. Cultivated and sandy fields; rare; 4-6. Annual, leaves broadly oblong, flowers subsessile, axillary. Ill, i. Foot of San Eoque, Wk. C. tricolor L. Fields ; common, sometimes abundant ; 4-6. Flowers bright blue, with white tube and yellow throat, sepals hirsute, in a cylindrical tube at base, spreading above, occasionally spathulate. A very pale-flowered form occurs, quite distinct from next. I. Eare, near Naval Hospital, K. II. ! III. ! C. meonantJnis Hoffm. & Link. Similar places, also on rough hills ; common, often a field full ; 4-5. Flowers smaller, pale lilac, not blue, sepals subglabrous, in a conical tube from base, not spreading above. The two species seldom grow together. III. i. Much commoner than last ! ii. and iii. ! 72 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD Calystegia sepium K. Br. Hedges and woods, perhaps often planted; frequent; 5-7. [I. Hedges at St. Bernard's, but not native on Eock !] III. ! Var. sylvestris Wk., flowers much larger, bracts very large and overlapping. I believe frequent, but have not distinguished it. III. ii. Damp places by K. Ancho near Algeciras, P. L. El Cobre!? C. Soldanella'R.'Br. Sand-dunes by sea; local; 4-6. III. ii. and iii. About Palmones, on both sides of River ! iCuscuta Eiyithymum L. On many low growing plants; rather frequent ; 5-7. The type has corolla tube much longer than the very broad short calyx segments, and does not, I believe, occur. fVar. angustata Engelm. has a very short corolla tube, the lobes and calyx segments triangular, longly acuminate. I. Upper and middle slopes ! III. i. Queen of Spain's Chair ! Carteian Hills ! Alcadeza ! ii. By Algeciras Cemetery ! Carnero Hills ! Common in mountains! Var. Kotschyi Engelm., differs from var. angustata only in smaller heads of closely sessile flowers, and stouter stem. I. On Bracliy podium pinnatmn, D. iVar. obtusata Engelm. has few-flowered glomerules, pedicels longer than calyx, and broadly ovate calyx and corolla lobes. III. ii. S. de Luna in Los Barrios district, Niks. BORAGINACE^. Hdiotropium europcsum L. Dry cultivated fields, roadsides, and waste places; frequent; 6-11. Var. temdflora Guss. is the commoner form in the province, and may be ours. I. Lighthouse! By Kennels ! Above Devil's Gap ! II. Forms large beds, K. I have not seen it there at all. III. i. By Francia's Farm! Near Bond's Farm ! Rail near San Roque Station ! ii. Algeciras Station, &c. ! iii. Palmones Village ! H. siqnnum L. Similar places, often on dried mud; rare? flowers too late for my observations of its distribution ; 6-7. III. i. In San Roque ! Roadside beyond Francia's Farm ! iii. About Los Barrios Station ! Cerinthe major L. Roadsides and fallow fields ; abundant ; 2-5. The type has yellowish-green bracts, and either yellowish or purple flowers, and is the sylvan form. I. Above Ince's ! Sunnyside Steps ! North Front, i^rere. II. ! III. i. Cork Woods, &c. ! ii. and iii. rare ! Var. purpurascens Boiss. Bracts as well as flowers dark purple; much commoner than type in open ground. I. K. II. K. III. ! Anchusa calcarea Boiss. Sandy fields; rare; 3-6. Flowers small, violet purple, calyx shortly 5-fid. III. i. At Foot of San Roque, Boiss, Wk., D. Var. scaherrima Boiss. is much more setose. III. i. With type, ib. A. italica Retz. Fields; common; 4-5. I. Mediterranean Road, but more abundant on lower parts, K. A specimen in herb. A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 73 Balestrino, labelled Lithospervium purpiireo-coeruleiim, probably came from the Eock, where it does not now, I think, occur. III. ! Borago officinalis L. Eoadsides, fields, and waste places ; very common, sometimes in masses ; 2-5. Occasionally with white flowers. I.! II.! III.! Echium Pomponium L. [E. glomeratum Boiss.). Eocky places; rare ; 5-6. The largest of our species, racemes many, lateral, densely covered with yellowish flowers. E. flavum Desf. may not be distinct; Lemann's specimen is so labelled. My own closely resembles Jacquin's figure of E. altissimum [E. italicum L.), but that has usually a much laxer panicle. I. Mediterranean Steps! E. pusttdatum Sibth. & Sm. Chiefly sand-dunes ; rather frequent ; 5-6. Very hispid, leaves narrow, panicle often cylin- drical, but equally often much branched, corolla usually reddish purple. I. Eocky places on south and west slopes, Wk., K., D. III. i. Towards Pedrera ! Near Tunares! Punta Mala! Cork Wood Crags, &c. ! ii. Palmones Playazo ! iii. Palmones Village and Sands ! [E. italicum L., resembling E. Pomponium, but usually much more branched, and with pale bluish flowers, has been found by Perez Lara at Boca de Leon, just beyond our limits.] \E. maritimum Willd. Sandy ground ; rare ; 3-6. Distin- guished from all others but E, calycinum by its included stamens, and from that by its larger deeper purple flowers. I. Devil's Gap to Queen's Gate ! III. i. Punta Mala ! Sea sand at foot of S. Carbonera and of San Eoque, Wk. E. plantagineum L. Sandy fields; abundant; 4-6. Leaves almost silky, stem little asperous, flowers very large. Varies greatly in habit, either with a single, erect, simple or branched stem, radical leaves withered by time of flowering, or, more commonly, with a radical rosette of very broad leaves, with several prostrate lateral and one central erect stem. I. Eare upper, more frequent lower slopes ! North Front ! II. ! III. ! E. creticum L. Eocky slopes, very common on Eock ; 3-5. Like the erect form of last, but much more asperous, especially leaves. I. ! II. K., D. III. i. Sand-dunes at San Eoque, Wk. Alcadeza ! E. calycinum Viv. {E. parviflorum Moench.). Eoadsides and stony places ; common ; 2-4. Flowers small, pale blue, stamens all included. I. General, but chiefly south ! [Litliospermum purpureo-ccemleum Willd. was communicated to Kelaart, certainly in error. A specimen of AncJutsa italica is so labelled in herb. Balestrino.] L. fniticosiim L. Heathy places in woods and mountains ; frequent ; 1-6. Perez Lara says our variety is his prostratum, but the characters are very indefinite, and Debeaux only admits var. erectum Coss. I. Frere. III. i. Queen of Spain's Chair I Pine Wood Plains ! Cork Woods! ii. Mountains! [L. officinale L. Eecorded from the Eock by Kelaart, either in error or as a casual.] 74 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD L.arvenselj. Fields and roadsides ; rare; 3-5. III. i. Linea! Near San Eoque ! ii. Algeciras Station ! L. apulumYAhX. Dry hills ; frequent; 3-5. III. i. Carteian Hills ! Campamento Common ! Myosotis repens Don. (ill. palmtris Eoth. var. bcetica P. L.?). Marshes and streams; locally common; 4-6. Calyx tuhe conical, deeply cleft, teeth much longer than broad, flowers rather large. III. i. Near top of Chair ! ii. Mountains! iii. Guadacorte ! filf. lingulata Lehm. {M. ccespitosa Schultz). Similar places ; rare ; 4-6. No aerial creeping stolons, leaves narrower, flowers smaller, corolla limb not wider than length of tube. III. ii. Hills near San Bernabe ! f M. sicu/a Guss. Similar places; rare; 5-6. Annual, pedicels ebracteate, scarcely longer than calyx, usually ascending in fruit. III. iii. Boggy meadows near Palmones, Winkl. \M.maritima Hochst.? Marshes near sea; rare; 5. Perennial, base of stem woody, hairs on tubercles, flowers very large and pale. An Azorean species. III. ii. Near Algeciras, Eev. M. hispida Schl. {M. collina Eeichb.). Sandy places, or in woods ; frequent ; 3-5. III. i. Cork Woods ! ii. Palmones Playazo and Pinar ! iii. Palmones Sands ! M. versicolor Pers. Similar or more grassy places; occasional; 3-5. III. i. Andalucian racecourse. Pinar de los Bigotes ! Cork Woods ! ii. Algeciras Golf Links ! Palmones Playazo and Pinar! Near M. de la Torre! iii. Palmones Sands! M. intermedia Link. {M. arvensis Hill). Sandy fields; rare; 3-5. III. [i. Near Gibraltar, Jf. Not "at Gibraltar," as Debeaux says.] ii. Near Algeciras, Clem. [M. sylvatica Hoffm. Debeaux credits Kelaart with this from near San Eoque, but Kelaart only reports it from the neighbour- hood of Gibraltar, probably outside our limits.] Cynoglossum cheirifolium L. Eocky and stony places ; oc- casional ; 2-4. The corolla is described as rosy, fading to violet or blue, but the limb is permanently cream-coloured, the tube and throat processes deep maroon. I, Jews' Cemetei'y ! Ince's ! Willis's ! III. i. San Eoque, D. ii. Algeciras, D. C. clandestinum Desf. Dry hills and sandy places ; frequent ; 2-4. III. ! G.pictum Ait. Similar places; frequent; 3-5. I. In great abundance on middle parts, K. I have only seen a single specimen near Green's Lodge ! III. ! Omphalodes linifolia Pourr. Dry bushy places ; locally fre- quent ; 5. III. i. Cork Wood Crags ! Alcadeza Crags ! [Symphytum tuberosum L. was communicated to Kelaart from the Eock. Error or casual ?] S0LANACE.aB. Solanum nigrmn h. Eoadsides and waste places ; rare; 1-12. The black berries are the only constant feature ; leaf lobing, hairiness, and size and colour of flowers vary greatly. I. Sand- A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 75 pits! North Front! III. i. Cork Woods ! ii. El Cobre! Near M. de la Torre ! Var. miniatum Mert. & Koch. Much commoner. I. ! III. ! Kare in ii. Var. siiffnUicosum Moris is 3-5 ft. high, woody at base. A common var. in the province, but rare with us. I. Devil's Gap and elsewhere ! S.villosumJjQ,mk. Similar places ; occasional? 1-12. Softly pubescent, musk-scented, flowers twice as large, fruit orange yellow. I have not seen fruit, and believe my gatherings are villous form of *S'. nigrum. I. Common, K., D. Above Engineer Koad ! ? III. i. Almendral ! ? ii. Algeciras Station ! ? El Cobre ! ? with small violet flowers. >S. sodomcEum L. Waste sandy places, mostly near cottages ; frequent; 2-11, but chiefly 4-9. I. On lower, rarely upper Eock! II.! III. i. Linea ! Alcadeza ! Eare in Cork Woods! ii. and iii. ! On charcoal burnings in the mountains 1 S. Dulcamara L. By streams and sotos ; locally frequent ; 6-8. III. i. By Lajo from First Pine Wood to Almendral ! Arroyo Viejo ! Cork Woods! [Lycium europaum L. Hedges ; rare, probably nowhere native ; 3-5. Corolla narrowly funnel-shaped, pale purple. I. Devil's Gap! Below main road, near St. Bernard's! III. i. Near Francia's Farm !] [L. afrum L. Corolla broadly funnel-shaped, dark brownish purple. Hedges of gardens in Gibraltar, D.] Mandragora autumnalis Spreng. Fields and grassy places ; common ; 9-1. III. i. Carteian Hills ! Arroyo Viejo ! Almo- raima ! ii. Hills near Algeciras ! Datura Stramonium L. Waste sandy places, chiefly near buildings ; occasional ; 4-11. I. Sporadic, as on North Front ! III. i. Near San Eoque Station ! Campamento ! Above Almo- raima ! iii. Palmones Village ! fVar. chalybea Koch has stem, petioles and calyces violet, flowers bluish. I. In lower part of town, D. [D. Metel L. with entire leaves has occurred as a casual near the Landport, K., Pourr.] [D. arborea L. is only cultivated.] [Hyoscyamus niger L. has been found as a casual on Europa Flats, Frere.] H. albits L. Waste places; type rare; 1-12. I. Near Monkey's Cave ! Var. viajor P. L. has throat and stamens blackish purple, and is the commoner. I. Europa Flats ! Moorish Castle ! Near slaughter houses! Mediterranean Eoad! III. i. Alcadeza Crags ! ■■'-Nicotiana glauca Grab, is quite naturalised at Catalan Bay, Engineer Eoad, Dockyard, and at Puente Mayorga ! [Cestrum Parqui L'H6rit. is an occasional escape. A bush grows on Neutral Ground just beyond our lines, but may be destroyed by the new road !] 76 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD SCROPHULARI ACEiE . Vcrbascum thapsiforvie Schrad. Sandy ground ; locally frequent; 4-6. I.? Leaves apparently of this species on slopes above Catalan Bay ! ? III. i. Cork Woods near Soto Gordo ! ii. Palmones Playazo ! iii. Palmones Village ! V. sinuatum L. Sandy banks and roadsides ; frequent ; 6-9. I. ! II. ! III. ! With white flowers by railway near Guadacorte ! V. virgatwn With. Similar places; rare; 5-6. III. i. By river above Almoraima ! Lane to First Venta ! Near San Eoque, Brouss. iii. Palmones Village ! i Scrophularia laxiflora Lge. By streams and in woods; locally common ; 3-5. Cymes very lax, much more so than in S. aquatica, flowers often pale, leaves not auricled. I. Near Willis's and catchment above ! III. i. Cork Wood Sotos ! ii. Mountain valleys ! [S. lavigata Vahl. A specimen from the Waterfall, my 884, exactly matches one from Ball, from same neighbourhood, labelled thus, but with a sign of doubt. It may be a form of last with leaves acutely doubly serrate. It is a Moroccan species.] *S'. auriadata L. {S. suhverticillata Moris). Damp places in open ground, not in woods ; frequent ; 3-9. Stout, cymes very dense, very dark-flowered, in a narrow raceme, flowers con- tiguous, bracts very broad, deeply lacerate, also sepals ; leaves usually with one or two pairs of pinnae at base. II. ! III. ! f Var. minor Lge. (L. aquatica L. ?) has probably been con- founded with S. laxiflora, if it be really distinct. III. ii. Alge- ciras, Bev. S. samhiLcifolia L. {S. mellifera Vahl.). Koadsides and ditches ; rather common ; 2-4, I. Sandpits ! Above Alameda Parade ! Engineer Eoad ! III. i. and ii. ! S. canina L. Sandy ground and hedges ; rather frequent ; 3-5. The type (not recorded) has leaves much divided, with toothed lobes. Var. }nnnatifida Boiss. Leaves dentate or irregularly pinna- tifid. III. i. Alcadeza Plains ! Near Soto Gordo ! Near Majarambout Woods ! Cork Wood Crags ! Var. frutescons Boiss. More woody, leaves less cut. Var. hcntica Boiss. hardly differs. I. Schucht ! Rare, K. III. i. Fre- quent about San Roque, K. Alcadeza Plains! ii. Palmones Playazo! iii. Palmones Sands ! [Anarrhinum laxifl,oriim Boiss. was gathered by Reuter between San Roque and Grazalema, which is probably outside our limits. Debeaux cites it as " rocky hills of San Roque."] A. bellidifolium Desf. Bushy hills; rare; 5-6. I. Vo7i Martins. III. i. Sea-sand at foot of Fort S. Felipe, D. Pedrera, D. Alcadeza Crags ! [Antirrhinum Linkianum B. & R. is one of Gandoger's determinations from rocks in the neighbourhood of Gibraltar, Dasoi. Stem woody, leaves very broad, flowers bright purple. Probably an error.] A. majus L. Rough bushy places ; common on Rock, rare in A FLORA OF GIBRALTAR AND THE NRIGHBOURHOOn 77 Spain; 3-6. I.! III. i. Alcadeza Crags! Valley near Long Stables ! ii. Near Algeciras, N6e. Var. ramosissimum Wk. has many elongate flexuose branches; mixed with type and perhaps commoner. I. ! III. i. Near San Eoque, D. ■\A. tortuosum Bosc Similar places; frequent? 3-8. Much like last var., but wholly glabrous, leaves and sepals narrower. My only gathering for A. majus proved to be this. I. Both sides of Eock, Bent., Nilss, &c. Below Mediterranean Eoad ! A. Orontium L. Fields and heathy places ; common ; 3-5. Very variable in leaves and flowers. I. ! III. ! Var. grandiflorum Chav. has broad leaves and large flowers, often much branched. Common in Spain, rare on Eock. I. ! III. ! f Var. iMrviflorum Lge. has flowers scarcely longer than calyx. III. i. San Eoque, D. ChcBnorrhinuvi villosum Lge. Eocks ; locally frequent ; 5-6. I. ! III. i. Alcadeza Crags ! Var. 2^usillum Boiss. has very small leaves. I. Winhl. ! ■■'Linaria Cymbalaria L. On walls ; occasional, but very doubtfully native, about the town and at San Eoque ! L. cirrhosa Willd. Banks and rough places ; rare ; 5-7. Leaves narrow, sagittate, pedicels long, slender. Debeaux's notes after this species (Fl. p. 148), refer to L. lanigera, and possibly his San Eoque stations refer to the latter also. I. Brouss. ! III. i. About San Eoque, Brouss., Pourr. S. Carbonera, Nilss. L. spuria Mill. Similar places ; rare ; 6-9. Calyx segments broadly oval, subcordate. III. ii. Near Algeciras, Bev. Var. racemigera Lge. has very short pedicels and very small calyx segments. Perez Lara thinks it is synonymous with next. III. ii. Algeciras, Bev. L. lanigera Desf. Similar places ; rather frequent ; 6-10. Very near last, but calyx segments narrow lanceolate. I. Abun- dant near messroom of South Pavilion, K., herb. Balestrino!; not there now, I think. Behind the Mount, K. II. ! III. i. and ii. ! seldom seen by me in flower, but I think frequent. L. triphylla Mill. Cultivated fields ; rare ; 3-5. Annual, leaves large and broad, flowers whitish or yellowish, variegated with lilac. I. Brouss. ! III. i. At foot of San Eoque, D. [L. Glementei Haens. a perennial, 3 ft. high, flowers violet with yellow palate, has been reported from Gibraltar by Dasoi.] L. viscosa Dum. Sandy ground; common; 2-6. Tall, with large yellow flowers in a compact raceme, scarcely longer in fruit. Often misnamed L. spartea L., which has much longer pedicels in a lax raceme. I. Lem. ! ? I think this, but fruiting raceme elongate. III. i. and ii.! In profusion on mountains after a fire I L. pedunculata Spreng. Sand-dunes ; locally plentiful ; 3-5. I.! II.! III.! From Catalan Bay to Sandy Bay beyond Algeciras ! Flowers purple or yellow. L. amethystea Hoffm. & Link. Sandy and gravelly places ; locally common; 2-5. Flowers violet. TIL i. Cork Wood and Alcadeza Crags! 78 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD Var. albiflora Boiss. has petals white, spotted with violet. I. Chiefly east side ! III. i. Queen of Spain's Chair, rare ! L. Munhijana B. & E. Sandy places; occasional? 2-5. Very small, with yellow flowers, seeds with a thick wing. A species I believe to be this, frequent in the stations cited, has very dark seeds, with a dark, but scarcely thick wing. I have seen no other specimens, but all allied species have pale seeds, with a very pale diaphanous wing. III. i. Between Gibraltar and San Koque, Schousb. Beyond Second Pine Wood ! ? ii. Between M. de la Torre and Palmones Bridge ! ? iii. Palmones Sands ! ? [L. Haenseleri B. & E., like last, but seeds with a very pale thin wing, is reported by Dasoi from the neighbourhood, but not confirmed.] L. tristis Mill. Eocks ; locally frequent ; 3-6. Flowers varying from dingy yellow to reddish brown. I. ! III. i. Alca- deza Crags ! San Eoque, B. d B. L. melanantha B. & E. Eocks; rare; 3-6. Flowers almost black, but often drying quite pale, leaves narrow, \ line or less, calyx segments narrower. Scarcely distinct from last. III. ii. Waterfall Valley ! Digitalis imrpurea L. var. tomentosa Webb. Woods ; locally frequent ; 4-6. III. i. Almoraima ! Mountain Woods at San Eoque, D. ii. Mountains! \Lafuentea rotundifolia Lag. Bushy rocky places ; rare ; 4-5 ; not found recently. I. Brouss. Sibthorpia euroijcea L. Wet places in mountairrs ; locally common ; 5-9. III. i. Foot of Alcadeza Crags ! ii. Eeaching high up the slopes ! Veronica Gymbalaria Bod. Walls and stony places ; very common ; 1-4. I. ! III. ! V. agrestis L. Sandy fields; rare; 12-3. Flowers axillary, pale, calyx-lobes oval, capsule notch narrow, valves 4-6-seeded. I. North Front, near Cemetery ! III. ii. Algeciras Station ! \V. persica Pourr. Similar places; rare; 1-3. Much larger, flowers bright blue, capsule lobes broad, flattish, divaricate. III. ii. Palmones Playazo ! V. polita Fries. Grassy places ; rather rare ; 3-4. Flowers bright blue, calyx lobes broadly ovate, capsule notch broad, valves 7-10-seeded. III. i. Lajo Eiver bed ! Almoraima ! ii. Alge- ciras Station ! V. arvensis L. Similar places ; frequent ; 2-5. Flowers minute, blue, in bracteate racemes. I. North Front! III. i. First Pine Wood ! Cork Woods ! ii. x^lgeciras Station ! Mountains ! V. Anagallis L. Wet ditches ; frequent ; 4-6. A polymorphous species. Mr. Druce refers specimens from Lajo Valley to the segregate F. aqiiatica Bernh., known by its very spreading pedi- cels and broad capsules, not narrowed below. Type V. Anagallis L. has ascending pedicels and capsules narrowed below. I cannot say which form prevails. III. ! V. anagalloides Guss. Similar places ; frequent in the A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 79 province ; 4-6. Leaves and sepals narrowex', capsule narrower, attenuate above. I have not identified it. II. D. Eufragia viscosa Benth. Grassy places ; common ; 4-5. I. South and west slopes, It., D. Signal Station! II. ! III.! Trixago apula Stev. Usually in drier places ; common ; 4-5. Flowers lavender and white. I. Breakneck Battery ! Moorish Wall! II.! III.! f Var. lutea Lge. has bright yellow flowers, known from Eufragia by its short broad sepals. L East slopes Middle Hill ! III. i. Eocky slopes of San Eoque, Wk., K., D. \Bartsia aspera Lge. Dry rocky places ; locally frequent ; 8-9. A Portuguese species, probably only a very scabrid form of the Pyrenean B. spicata Ram. III. ii. Tops of mountains ! whence Reverchon records it. Odontites tenuifolia Don var. australis Boiss. Dry bushy places ; rare ? 7-9. Very slender, flowers small, unilateral. The var. is a more glabrous form. III. ii. S. de Palma, Bev. Orobanchace^. In this difficult order I have, as a rule, only mentioned localities from which specimens have been named for me by Prof. Beck, as well as citing those already recorded. It might be misleading to give their apparent distribution, though several species are probably quite common. Orohanche cernua Loefl. Apparently rare; 4-6. Corolla bluish, much deflexed, much constricted below mouth. III. ii. Near Algeciras, Winkl. W- caryophyllacea Sm. (0. Galii Vauch). Apparently rare ; 3-5. Stout, corolla large and broad, wholly brown, stigmas brownish or violet. III. ii. S. de Palma in Los Barrios District, Rev. if. macroglossa Beck has corolla square in lateral aspect. III. i. Cork Wood Crags! (my 1669, pars, teste Beck). Con- founded with 0. gracilis. i 0. Eap2t)n-Ge7ust(s ThuiW. Appears rare; 3-5. Larger than 0. gracilis, with larger wider corolla, more uniformly brownish rose, not shining within ? stigmas yellow. III. ii. S. de Palma, Wiiikl., Bev. 0. gracilis Sm. (0. cruenta Bertol.). Rather frequent, at least locally ; 3-6. Usually large, stout, corolla dull greenish yellow outside at base, reddish towards mouth, deep shining moroon inside, stigmas yellow. III. i. Queen of Spain's Chair ! (teste Beck), ii. Frequent at the Waterfall ! {teste Beck). ff. polyantha Beck has smaller flowers in a longer spike. III. i. Cork Wood Crags! (my 1669, pars, teste Beck). fVar. Spr miner i Beck (0. reticulata Beck, non Wallr.). Corolla larger, upper lip with broader lobes. III. Near Gibraltar, Beck. ii. Algeciras, Beck. 0. foetida Poir. Either this species or the next often frequent in cornfields ; 4-5. Corolla deep blackish crimson, 7-12 lines, filaments pilose below, inserted li-3J lines above base, stigmas described as yellow, but Prof. Beck thus names my specimens 80 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD with them deep red. I. K., D. III. i. San Eoque, K., D. ii. Palmones Playazo, on Ononis pinnatal (my 1920, teste Beck). S. de Palma and S. de Luna, Beck. tf.^ws«7/a Beck is a reduced form. III. i. Second Pine "Wood ! (my 1978, teste Beck). to. sanguinea Presl. (0. crinita Viv.) ; 4-5. Like last, but corolla shorter, filaments subglabrous, inserted much nearer base, stigmas red. I. Bruce's Farm ! (my 1958, teste Beck). On Lotus, Salzm. III. i. Algeciras, Beck. [0. reticulata Wallr., recorded from the Eock by Lemann, and from San Roque by Dautez, is probably an error for 0. gracilis. It is not known in S. Spain.] 0. crenata Forsk. (0. pruinosa Lap., 0. speciosa DC). On peas and beans ; common, often a field full ; 4-5. Tall, very stout, corolla large, lavender or bluish and white. I. Sand- pits! III.! 0. densi flora Salzm. Rare, or passed over for one of the other yellow-flowered species ; 4-5. Flowers yellow, in a very dense spike, corolla 5-8 lines, bracts as long as corolla tube, filaments auriculately lobed at base. III. ii. Near Palmones and Alge- ciras, Hack., Winkl. {teste Beck). Algeciras, Bev. ■\0.mauretanica'Beck. Rare; 5-6. Near last, but spike longer and laxer, flowers much larger, bracts as long as corolla, filaments not lobed at base. Usually wholly bright yellow, or with brown bracts and sepals, and corolla streaked brown. Stigmas yellow or pink. III. i. Between Sprague's Farm and San Roque ! (my 1985, 1986, teste Beck). A specimen from I. Devil's Gap (my 1939) may be the same. [0. pubescens D'Urv. [0. versicolor Schultz). A specimen at Kew, from Algeciras, so labelled by Winkler, is unnoticed by Beck, and is probably one of the other yellow- flowered species.] to. loricata Reichb. Rare; 5-6. Flowers yellow, filaments hairy at base, very glandular at apex, bracts longer than corolla, calyx segments deeply divided, stigmas pink. III. ii. Algeciras, Bev. {teste Beck). \0. Picridis F. Schultz. Apparently rare; 5-6. Very near last, but filaments glabrous at apex, bracts as long as corolla, calyx segments bidentate to middle or less. I. Near Breakneck ! (my 2118, teste Beck). A form near O.Boissieri Reichb. f., which is a step towards 0. ametJnjstea Thuill. f Var. CarotcB Beck is a dense flowered form, with less glandular- pilose reddish violet bracts, and subglabrous corolla with violet veins. I. Gibraltar, Beck. III. ii. Algeciras, Beck. 0. minor Sutt. Occasional ? 5-6. A most variable species. I. West slopes, K., D. III. ii. Algeciras, Bev. {teste Beck). f. concolor Beck (0. concolor Duby) is a wholly yellow form. III. iii. Sands at Guadacorte ! (my 1913, teste Beck). [Phelipcea ccerulea C. A. Mey (0. ccsrulea Vill.) recorded by Kelaart from the Rock is unknown in Spain.] [P. ramosa C. A. Mey. (0. ramosa L.) also recorded from Gibraltar by Kelaart is not known in the South of Spain.] A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 81 P. Muteli 'P.^chviltz. Apparently frequent ; 3-5. I.Gibraltar, Beck. III. i. San Eoque, Beck. ii. Algeciras, Beck. f P. nana Keichb. f. Rare? 4-5. Like last, but corolla shorter, segments of lip more acute, tube less constricted in middle, III. i. Cork Wood Crags ! (my 1736, teste Beck). Labiate. Lavandula Stcechas L. Dry bushy hills ; rare in Gibraltar, abundant in Spain ; 2-6. Leaves quite entire. I. South and west slopes, Tournef., ScJiott, A'., D. III. ! \L. dentata L. Similar places ; common on Eock, rare in Spain; 11-4. Leaves dentate. I. Chiefly above Alameda and above Levant, but frequent elsewhere ! A whitish-flowered form occurs near Jews' Cemetery! III. i. San Roque, D. '? Not confirmed. L. multifida L. Similar places ; common on Rock, rare in Spain; 11-4. Leaves pinnatifid. A white-flowered form is found. I. ! III. i. Near San Roque, Boiss., Pourr. Mentha rotundifolia L. Damp grassy places ; very common ; 6-8. I. North Front, K. A few plants above Devil's Gap ! II. ! III. ! f Var. macrostachya W.-Dod {M. macrostachya Ten.) has shorter denser spikes. III. ii. Algeciras, Rev. \M. Bauhini Ten. Similar places; rare; 6-8. Taller, less tomentose, spikes very dense, short and acuminate, many in a lax panicle. Perhaps only a variety of last. III. ii. Algeciras, Bev. \M. aquatica L. ? Similar places or in water; rare; 7-9. Debeaux throws doubt on the occurrence of this species, and Perez Lara does not record it, but I have seen leaves of what appears to be a glabrous form in the stations cited. III. i. In the Lajo at First Pine Wood ! ? ii. Near Arroyo Gaba! ? Slopes south of Algeciras ! ? iii. In a copse at Guadacorte ! ? M. Pulegium L. var. villosa Benth. Grassy places ; very common; 6-9. Young leaves and shoots usually quite glabrous, but flowering stems almost always very tomentellous. I. North Front, K, &c., not there now. II.! III.! Lycopus europceiis L. Streams and marshes ; locally rather frequent ; 6-9. III. i. Cork Wood Sotos ! ii. Mountains ! iii. Guadacorte ! Origanum virens Hoffm. & Link. Dry banks ; rare ; 6-8. III. ii. S. de Palma, Rev. 0. compactnvi Benth. Dry places ; rare ; 6-8. Differs from last in narrower more glandular bracts, and perhaps only varietally distinct. III. i. Road to Malaga Gardens ! Between San Roque and First Pine Wood! Thymus hirtus Willd. Rocks and walls ; rare ? 5-7. ]\Iuch confused with next, and Mr. A. B. Jackson is of opinion that all he has seen so labelled from the Rock belongs to that species. Willkomm suggests that it may be a variety of T. vulgaris. The distribution of our species requires further investigation. I. Boiss., D. III. i. Carteian Hills, K. ii. Waterfall Valley? not seen in flower. Journal of Botany, June, 1914. [Supplement] g 82 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD iT. diffusus Salzm. Similar places; common on the Rock; 5-7. Differs from last chiefly in narrower floral leaves, though in Masson's specimen, cited as typical, as well as in all I have examined, they are twice as broad as cauline. I. ! ]T. vulgaris L. Similar places and rough ground ; rare? 5-7. Cauline leaves much broader and flowers pedicelled. I. Gaudi- chaud. III. i. Carteian Hills, Von Martins, K., and others. i Var. capitatns Wk. has capitate inflorescence. I. Rev. My 917 from Moorish Wall near Middle Gate closely resembles this, but Mr. Jackson thinks it is only a form of last with lax branches and broader leaves. Coridothymns capitatns Reichb. f. Dry hills; locally abun- dant ; 5-9. I. Gaudichaud. III. i. North and West of San Roque ! ]Micromeria grceca Benth. Dry bushy places ; frequent; 4-6. Only the variety is recorded by Debeaux, though the type seems equally frequent. I. ! III. i. and ii. ! Var. latifolia Boiss. has leaves as broad as those of next. I. ! III. i. and ii. ! M. nervosa 'Q&nih. Similar places; rare; 5-6. Inflorescence much denser, calyx very villous, teeth more spreading. III. i. By Pedrera, D. Satureiainodora Ssdzm. Dry heathy hills ; rare; 6-7. Rather like Coridothymns, but much more lax and straggling, with much laxer inflorescence. III. i. Top of Queen of Spain's Chair ! ii. Cuartel de las Corzas, Laguna. S. de Palma, Bev. {Calamintha Nepeta Savi {Melissa Nepieta) is recorded by Boissier and Kelaart from Gibraltar. It differs from next, which is often mistaken for it, in its subequal calyx teeth, throat hairs exserted, and smaller corolla.] C. menthcBfolia Host. {G. officinalis Benth.) var. hcetica Ball. Similar places; locally abundant; 6-12. Lower calyx teeth much the longest, thi'oat hairs included. The variety is more densely white tomentose, and more branched, with larger corolla, but scarcely differs from type, which is abundant elsewhere in the province. I. ! III. i. Cork Woods ! ii. Waterfall Valley ! G. Clinopodium Benth. Woods and bushy places ; rather rare; 5-6. Our form seems to be var. pteroceplmla P. L., with long plumose calyx. III. i. Almoraima Soto ! ii. Waterfall Valley! Melissa officinalis L. var. villosa Boiss. Similar places ; locally frequent ; 6-7. The variety is a more villous form. III. i. Almoraima ! ? leaves only. ii. Waterfall Valley ! Bosmarinus officinalis L. Bushy and rocky places; rather rare; 11-5. I. Mediterranean Steps and Road t Near Signal Station ! III. i. Cork Wood Crags ! Salvia triloba L. t var. calpeana Daut. & Deb. Rocky bushy places; very rare; 5-6. Shrub, 5-6 ft., leaves narrow, entire, wrinkled, some (very few in the only bush I have seen) with a pair of leaflets at base. Differs from type in being twice as lai'ge, with longer laxer spikes, and broader much less tomentose leaves. The entire-leaved form, var. integrifolia Rev., hardly difi'ers from A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 83 S. officinalis, except in being shorter, with shorter, broader, less acuminate calyx teeth. I. About Ince's, apparently cultivated ! Dautez describes the station as "rocks on west slopes," and Debeaux adds, " the indigeneity of this plant in the rocky and elevated ravines could not be doubted," so it doubtless occurs elsewhere, though I have repeatedly searched in vain for it. Colmeiro thinks it not native. {S. officinalis L., S. hispanica L., and S. viridis L., are reported by Kelaart and Gaudichaud from the Eock. None is probable but S. lavandidcefolia may have occurred formerly.] [S. rotunclifolia is mentioned by Kelaart as found by San Eoque road. There are three species of that name, none of them Spanish.] f *S. Sclarea L. Eough slopes ; rare ; 5-6. Leaves broad, sub- cordate, bracts very broad, strongly veined, flowers pale blue. III. ii. Near a cottage in valley above Frayle Bay! >S. tingitana Etth. Bushy places ; rare ; 5-6. III. Neigh- bourhood of Gibraltar, Rouy. S. bicolor Desf. Banks of streams ; locally frequent ; 5-6. Very showy, often 3-4 ft. high, with large lavender flowers. [I. Communicated to Kelaart.] III. i. By Cagancha ! Arroyo Viejo ! Near Malaga Gardens ! Towards Alcadeza Crags ! iii. Mr. Patron informs me that it occurs with white flowers behind Guadacorte. S. hullata Vahl. (S. hcetica Boiss.). Dry banks ; rare? 4-6. Leaves rather large, ovate lanceolate, buUate, flowers fuscous red. III. i. Cork Woods, Boiss., Schousb. San Eoque Eoad, K. Plentiful near i\.lmendral, K. S. Verbenacah. Dry banks; common; 12-6. Corolla usually pale, rarely dark blue, much longer than calyx, upper lip much arched. Inflorescence very glandular, often short and dense. I. Chiefly lower Eock ! III. i. and ii. ! Var. 2>rcBC0x Lge. (S. clandestina L.). Debeaux evidently uses this name for the next species, but according to Mr. Pugsley (Journ. Bot. 1908, p. 144) it stands for dwarf forms of S. Ver- benaca. III. i. and ii. ! f S. horminoides Pourr. (non G. & G.). Similar places ; com- mon ; 12-6. Always tall and eglandular, corolla much darker and smaller than last, upper lip scarcely arched. III. i. and ii. ! Nepeta Uiberosa L. Eough bushy places ; occasional ; 5-7. Floral leaves and bracts quite herbaceous, not pellucid, reticulate- veined. I. By wall from Middle Gate to Signal Station ! Near Farringdon's ! Mediterranean Steps, a single plant ! III. i. Malaga Gardens ! N. reticulata Desf. Similar places ; rare or erx'or ? 7-8. Near last, but bracts submembranous, pellucid. I. Clem. '? N. Apulei Ucr. Similar places ; rare ; 4-6. Much more glabrous, floral leaves and bracts narrow, herbaceous, nerves strong, not reticulate. III. i. Near San Eoque, D. Lamium amplexicaule L. Chiefly in vegetable gardens ; rather 0 2 84 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD frequent ; 2-5. I. Everywliere, D., where I have not seen it. III. i. Cork Woods, far from cultivation ! ii. and iii. ! Pelayo! L. flexuosum Ten. Woods ; rather rare ; 4-6. Closely resembling L. album, but lower lip of corolla with one, instead of 2-3 teeth, anthers glabrous, not bearded. III. i. About Almo- raima! ii. Waterfall Valley ! Stachys cjermanica L. var. lusitanica P. L. Bushy hills ; very common ; 4-6. The local form of this polymorphous species does not vary much. It differs from type in its oblong cordate leaves, obtusely crenate, indumentum more tomentose, less silky, and flowers whitish, not purple. I. Above Engineer's Koad ! III. ! Var. interrupta Rouy, with whorls more distant than in type, seems hardly worth distinguishing. It is mixed with type every- where. III. ! S. circinnata L'Herit. Rocks ; locally rather frequent ; 4-5. I. ! Not confined to south and west, as Debeaux says. S. arvensis L. Fields and bushy ground; very abundant ; 1-5. Occasionally with pure white flowers. III. ! S. hirta L. Chiefly in fallow fields ; abundant ; 4-6. I. Near Jews' Cemetery ! Mediterranean Steps ! A casual on Line Wall ! North Front, Frere. II. ! III. ! Betonica algeriensis De Noe. Mountain slopes ; locally common; 5-7. A good species, I think. Inflorescence sub- spicate rather than capitate, flowers much smaller and paler than in B. officinalis, dingy pink, bracts and sepals often glandular punctate. III. i. Queen of Spain's Chair ! ii. Mountains ! Ballota hirsuta Benth. Eough bushy places; rare; 5-6. I. Roadside above Willis's ! South and west slopes, D. Phlomis Herba-venti L. Bushy places ; locally common ; 5-6. III. i. Roadsides and fields towards S. Lorca ! ii. Woods behind Algeciras, K. P. purpurea L. Bushy places; abundant; 3-6. I.I III.! With pure white flowers on the Rock and near M. de la Torre ! iP.fruticosaJ-i. Similar places; now extinct? 3-6. I. Tournef., Ga2idicJimid, K. P. Lychnitis L. Dry and stony places ; rare ; 5-6. Flowers yellow. III. i. South slopes of San Roque, D. Marrubium vulgare L. Waste places near buildings ; rather frequent ; 4-7. I. Below Signal Station ! Windmill Hill ! Catalan Bay ! III. ! Sideritis scordioides L. var. Cavanillesii Wk. Rocky places ; rare ; 4-7. Near next, differing mainly in calyx teeth with long spines, often spreading. I. Masson, Martins, Link., teste Wk. Kelaart and others record the next species under this name. ]-S. arborescens Salzm. Rough bushy places ; occasional ; 4-5. I. Levant ! Near Queen's Gate ! Engineer Road ! Described as 4-5 ft. high ; I have never seen it more than 2-3 ft. S. a7igustifolia Lam. Similar places; rare; 5-6. Much more slender and elongate, less shrubby, with a more spinose calyx and narrow entire leaves. III. i. Plain below Malaga Gardens ! a very glabrous form. A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 85 S. romana L. Dry and rocky places; rare; 4-5, Annual, with white flowers, floral leaves like cauline. III. i. S. Carbonera and San Eoque, D. Cleonia lusitanica Lt. Eough slopes; occasional; 5-6. III. i. Beyond Pedrera ! Path to First Pine Wood ! Pine Wood Plains ! Alcadeza Crags ! ii. Algeciras, Frere. Prunella vulgaris L. Damp marshy places ; frequent ; 4-6. III. i. Almoraima Soto ! By Lajo ! ii. Mountains ! Carnero Hills ! iii. Guadacorte Marsh ! \Prasium viajus L. Eocky places and old walls ; very com- mon on Eock, rather rare in Spain; 4-6. I. ! III. i. At Carteia! ii. Old walls at Algeciras ! Carnero Hills ! Ajuga Iva Schreb. Dry hills ; rather rare ; 3-7. III. i. Carteian Hills ! ii. Near Sandy Bay ! iii. Guadacorte ! Var. i^seudo-Iva Benth. with yellowish, not rose flowers, and more revolute leaves. Much commoner than type, and usually cleistogamous. I. Near Willis's ! III. i. and ii. ! j-Teucriuvi campamdatmn L. Eare ; 5-7. Leaves sub- bipinnatifid. III. ii. Near Algeciras, Nee. T. fruticans L. Bushy places ; chiefly in woods ; common ; 2-5. I. ! III. i. and ii. ! fVar. latifolium Eouy, with larger broader leaves, subvillous above. I. Wk., K, D. III. i. S. Carbonera, D. ii. S. de Palma, Bev. fVar. roHmdifolium Daut. & Deb. Leaves shorter, broader, half the size, roundish ovate, green above. I. III., i. and ii. Same collectors. \T. hracteatum Desf. Dry grassy hills; rare; 6-7. Whorls several-flow^ered, not secund, bracts ovate lanceolate, longly petiolate, calyx segments subequal, flowers pink. A Moroccan species, reported from Tarifa according to Ball, but not recently found. III. ii. Eidge between Carnero and Pelayo!, dwarf specimens in an exposed situation. T. Scorodonia L. Bushy and rocky places in mountains ; locally common ; 5-7. The type is subglabrous and eglandular. III. ii. El Cobre, &c. ! Other examples from Waterfall Valley have a more or less glandular-ciliate calyx, but are best under type. Var. hcBticum P. L. Inflorescence very glandular, corolla tube shorter. I. Brouss. !, Salzm., Durand, dx. III. i. Queen of Spain's Chair ! Alcadeza Crags ! ii. Mountains ! Var. pseudo- Scorodonia P. L. Leaves longer, less deltoid, whitish-tomentose beneath. I. Eare, K. III. i. Queen of Spain's Chair ! T. resupinatum Desf. Cultivated fields ; frequent or common ; 5-6. Leaves dentate or subentire, corolla resupinate. III. i. and ii. ! T. scordioides Schreb. Marshy spots ; very local ; 7-8. III. iii. Palmones Sands ! Thicket near Guadacorte ! \T. lucidum L. Eough slopes; rare; 6-7. Aspect of T. Cha- mcedrys L., but quite glabrous, with shining leaves. I.'? Between Old Man's Garden and Engineer's Eoad, Bobinson, Not confirmed. 86 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD T. Folium L. Eough bushy places ; locally common ; 4-7. I. ! III. i. Cork Wood Crags ! Near Malaga Gardens ! Var. cmreoforme Rouy has yellowish tomentum. Debeaux says he has submitted specimens from the Rock to M. Rouy, who says they are not T. aureum Schreb., as at first supposed. I. Near Middle Gate of Charles the Fifth's Wall ! Vbrbenace^. Verbena officinalis L. Waste places; frequent; 5-10. I. Not common, K. II. Frequent, K. III.! V. supina L. Similar places ; rare ; 4-6. II. D. III. i. Sand desert, D. ii. One or two places in Algeciras ! Vitex Agmis-castus h. Sandy ground; occasional; 7-8. II.! III. i. Near Agua Mayorga, D., Boiss. Lane to First Venta ! Almoraima I Towards S. Lorca ! ii. Near Algeciras and El Saladillo! Arroyo Gaba! ACANTHACE^. Acanthus mollis L. Bushy places ; abundant on dry lime- stone on the Rock, and by watercourses in Spain, very rarely dry places ; 4-6. I. ! III. ! Long Stables Ravine and Alcadeza Crags are the only dry spots in which I have seen it. Lbntibulariace^. Pinguiculalusitanicalj. Stream beds; locally frequent; 4-7. III. i. East slopes Queen of Spain's Chair ! ii. Mountains ! Primulace^. Coris monspeliensis L. Sandy places; locally frequent; 3-6. III. i. Cork Wood Crags ! Alcadeza Crags ! S. Carbonera and San Roque, D. Lysimachia Epliemerum L. Damp bushy places ; locally fre- quent ; 6-7. III. i. Almoraima Soto ! iii. Guadacorte Marsh ! Asterolinum stellatum Hoffm. & Link. Dry hills and moun- tains ; common ; 3-4. III. i. Queen of Spain's Chair ! Bond's Farm ! About San Roque ! Cork Woods ! ii. Mountains I ■\Centimculits minimus L. Damp grassy places; rare; 4-6. III. i. Near San Roque, Ball ! ii. Near Algeciras, Nilss. Anagallis crassifolia Thore. Marshy places; locally frequent; 4-6. Flowers white. III. i. Near top of Queen of Spain's Chair ! Alcadeza Crags ! ii. Mountains ! A. arvensis L. Dry hills and fields, tilled and native; com- mon; 2-5. Flowers red. I. A'. II.! III.! With pale pink flowers (var. carnea Schr.) behind Algeciras engine shed ! A. ccerulea Lamk. Similar places ; abundant; 2-5. Flowers blue. Varies greatly in size, the small-flowered forms may be A. parvifiora Hoffm. & Link. I. ! II. ! III. ! Var. latifolia Lge. is larger, stouter, leaves broadly ovate, subcordate, semiamplexicaul, calyx segments longer and broader. Frequent on Rock. Perez Lara says it is commoner than type in province, but I have not found it so. I. ! III. ! A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 87 fii. i^latiipliylla Baudo. Similar places; rare; 5. Annual, like last, but flowers 1\ in. diameter, leaves 1 in. broad. III. i. Rocky grassy places on south slopes of San Roque, D. A, linifolia L. Sandy places ; locally frequent ; 4-6. Perennial, leaves linear or narrow lanceolate, flowers very large. I. Masson, Finlaij\, B. cO B. II. D. III. i. Alcadeza Crags and Plain! Cork Wood Crags ! ii. Palmones Playazo! iii. Palmones Sands ! Samolm Valerandi L. Wet places; frequent; 4-9. I. Caves at Europa Point, K. Rocks over North Front, D. II. ! III. ! Plumb AGIN ACE^ . Armeria macro2)lujlla B. & R. {A. bcetica var. stenophylla, Boiss.). Heathy places; locally frequent; 4-6. Leaves long, very slender, in dense tufts, calyx lobes truncate, mucronate. I.? Finlajjl Willcl, teste Wk., Masson. III. i. Cork Woods! Pine Wood Plains ! A. bcetica Boiss. Grassy places; locally frequent; 3-6. Leaves shorter and broader, calyx lobes acuminate, cuspidate. III. i. Bond's Farm ! Linea and Campamento, Frere. ii. and iii. ! Var. hirta Boiss. has leaves and calyx ribs more hirtulous. II. or III. i. Sands near Gibraltar, Willd., D. ii. Sandy Bay I ? Statice ferulacea L. Salt marshes ; locally frequent ; 5-7. Leafless, flowers at ends of branches, outer bracts and calyx longly aristate. III. i. Rare on Bond's Farm ! ii. Algeciras, Rev. iii. Guadarranque Marshes ! Aguacorte ! S. diffusa Pourr. Similar places ; rare ; 6-7. Like last, but flowers fewer, at base of branches, bracts and calyx not longly aristate. III. ii. Near Algeciras, Nilss. iii. Palmones, Nilss. S. virgata Willd. Similar places ; rare or error ? 6-10. Leaves narrow spathulate, panicle with many barren branches. I. Sands and rocks by shore, Lagima? Possibly the next mistaken for it. S. simthulata Desf. var. emarginata Boiss. Rocks by shore ; locally abundant ; 5-6. Dense tufts of rigid spathulate emarginate leaves, no sterile branches. I. From Europa Point to Monkey's Cave I Beyond Catalan Bay ! III. ii. Algeciras, Bev. S. hjchnidifolia Gir. Muddy salt marshes ; locally common ; 5-6. Radical leaves few, large, oblanceolate. III. i. Above Guadarranque Ferry ! ii. Palmones Playazo! iii. Guadarranque Marshes ! Recorded by Nee as S. ovalifolia Poir. ? S. sinuata L. Sandy slopes and rocks near sea ; locally common ; 4-6. I. Europa Point to Governor's Cottage ! III. i. Sands at foot of S. Carbonera, Wk. ii. Algeciras, Bev. Plantaginace^. Plantago Psyllium L. Dry slopes and fields ; very common ; 3-5. The common form is var. dcntifolia Wk., with broader leaves, with 2-3 rather long teeth each side, but I think the type with entire or denticulate leaves also occurs. I. ! II. ! III. ! P. amplexicaulis Cav. Similar places ; rare ; 3-6 ; Annual, stems 1-4 in., leaves mostly basal, linear lanceolate, petioles short. 88 A FLORA OP GIBRALTAR AND THE NElGHBOtFRHOOD broadly amplexicaul, and large subglabrous flowers. III. ii. Algeciras, Winhl. [P. Laflingii L. was reported in error to Kelaart from the Kock.] P. Lagopus L. Similar places ; frequent ; 4-6. Like P. lanceolatum but annual, with very longly villous calyx. Var. lusitanica Ball differs chiefly in the quite unimportant character of taller scapes, and is inseparable from the type. I. Windmill Hill!, and elsewhere in the south, De6. III. i. San Eoque, D. ii. Near Algeciras, Nilss. iii. Near Los Barrios Station ! Var. vaginata W.-Dod (P. vaginata Vent. Jard. Cels. t. 29) has a long stem sometimes 1 ft. high. It is as common as type or var. lusitanica, and is connected by every intermediate. I.! III. ! P. lanceolata L. Grassy places ; frequent ? 3-5. Perennial, corolla and calyx glabrous. II.! III. i. Opposite Francia's Farm and elsewhere ! ii. About Algeciras ! Carnero Hills ! The form is usually var. erioi)hylla Dcsne., but I have seen type. P. Bellardi All. Dry sandy places ; very common : 4-5. A low annual, with short thick spikes and grey pubescent leaves. I. Willis's ! South and west slopes, K., D. III. i. ! P. Serraria L. Eather damp clayey soil ; abundant ; 4-5. Leaves bright green, glabrous, remotely coarsely dentate, spikes long, cylindrical. I. Windmill Hill ! II. ! '? III. ! f P. maritima L, ? Salt marshes ; rare ; 4-5. Leaves very narrow, usually entire, capsule 2-seeded. III. iii. My 2060 from near Los Barrios Station may be this, but the leaves are rather broad and have one or two teeth, the capsule is 2-seeded, but this sometimes occurs in P. Serraria. P. Coronopits L. Sandy ground ; abundant ; 1-12. Fairly constant in character in our region. I. ! II. ! III. ! Var. crithmifolia Wk. (P. Geratophyllum Hoffm. & Link.) is larger, with bipinnatifid leaves, and a broader rachis. I. Catalan Bay ! Sentry Fence ! Governor's Cottage ! P. major L. Chiefly in vegetable gardens ; occasional ; 3-9. I. K. III. ! [Nyctaginace^.] \Mirabilis Jalapa L. is more or less established above Devil's Gap, at Sandpits, and in gardens.] Amaranthace^. Amaranthus albus L. Eoadsides and gardens ; occasional ; 6-9. Pale green, with sessile axillary spiny clusters. III. i. Near Punta Mala ! Eailway near Second Venta ! A weed of cultivation elsewhere! A. Blitmn L. Similar places; rare; 5-9. I.? "Gibraltar," Brouss. without precise locally, but perhaps not on the Eock. A. cJilorostachys Willd. Waste and cultivated soil; rare; 12-5. Dark green, with dense terminal and lateral spikes, floral bracts spinose. III. i. Gardens at First Venta! ii. Algeciras Station ! A. defiexus L. Similar places; frequent; 1-12. Prostrate, with short axillary and terminal spikes, bracts muticous. Leaves A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 89 often blotched with white or black as in A. Blituvi. About the Town and at South Barracks ! II. ! III. i. River bed at Almo- raima ! Campamento ! ii. Algeciras Station, &c. ! \Acliyranthes argcntca Lamk. Grassy bushy places ; local ; 3-6. I. Above Alameda Parade ! Below Mediterranean Tunnels ! Monkey's Cave ! Dockyard ! Sandpits ! Above Catalan Bay ! ■''Alternanthera Achyraniha R. Br. By walls and waste places; rare ; 7-10. Clusters axillary, silvery white. I. Haensl. III. ii. Algeciras, Rev. '■•'\Pupalia atroimrpurca Moq.-Tand. Similar places ; rare ; 5-6 '? Inflorescence spicate, in distant globose, woolly spiny clusters. III. ii. Algeciras, Bev. Chenopodiace^, Salsola vermiculata L. Salt marshes ; locally frequent ; 6-9. Shrubby, with white stems. A glabrous form, probably var. microphylla Moq.-Tand., but erect, not prostrate. III. iii. Gua- darranque Marshes ! S. Kali Ten. In deep sand on or near the shore ; very com- mon ; 6-9. I. Near Lighthouse ! II. ! III. ! Var. rosea Moq.-Tand. Glabrous, purplish, wings of perianth rosy. II. K., D. III. i. Sand Desert, K., D. S.Sodah. Sandy places near sea; rare; 7-9. 1. Gaudichaud. III. ii. Marsh at Palmones Playazo ! iii. Guadarranque Marshes ! Halogeton sativus Moq.-Tand. Salt marshes ; rare ; 7-9. Like a Suceda, but with long, membranous, spathulate perianth segments. I. K. Suceda maritima Dum. Salt marshes and sands near sea; occasional; 7-10. I. By Inundation! Europa Point! II. K.,D. III. ii. Algeciras, K., D. iii. Palmones Marshes ! an erect form, subwoody below, but apparently annual. S. fnUicosa Forsk. Similar places; rare; 5-11. I. Rocks below Europa Point ! Probably the plant which Kelaart thought might be S. viaritima var. salsa Moq.-Tand., but quite suffruticose. [Salicornia fruticosa L. is recorded from Neutral Ground and Algeciras by Kelaart and Dautez, but almost certainly in mistake for the next.] Arthrocneimim macrostacliyum Mor. & Delp. Salt marshes ; locally abundant; 1-12. Each flower in a cup-shaped receptacle in one excavation of rachis, not each naked in a separate recep- tacle. II. K., D. ? III. ii. Palmones Playazo ! iii. Gua- darranque Marshes ! Varying greatly in size and colour. Young plants have the appearance of S. herbacea. Kochia scoparia Schrad. Salt marshes ; rare ; 9-10. III. ii. Near Algeciras, Nee. Obione porUdacoides Moq.-Tand. Salt marshes and waste places near sea; rather rare; 8-10. I. Reclamation Road! III. ii. and iii. Palmones River and Marshes ! Atrip)lex Halimus L. River banks and near sea ; local ; 8-9. III. i. By Lajo between the fords ! ii. Common on sandy shore hills at Algeciras, Bev. 90 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD A. hastata L. ? Eoadsides and waste places ; common; 7-9. Only seen in young state. I. Governor's Cottage and elsewhere ! II.! III.! A. angustifolia Sm. Similar places; rare; 7-11. III. i. Eailside about San Roque Station ! iii. Palmones Marshes ! ■'•\Bouhieva muUifida Moq.-Tand. Sandy places near sea; abundant, at least locally; 8-10. I. Eosia! North Front! II.! III.! Chenopodium ambrosioides L. Eoadsides and waste places ; rather frequent ; 6-11. The plant has a strong aromatic smell, and varies much in leaf cutting. I. Eosia ! Eeclamation Eoad ! II.! III. i. Almoraima Station! First Venta! Towards Bonel's Farm ! ii. Algeciras Station and elsewhere ! tVar. ■pinnatifida Wk. has pinnatitid leaves. III. iii. By Guadacorte Farm ! G. album L. fvar. pagamim Eeichb. Cultivated and waste places ; rather frequent ; 6-10. Leaves narrowed below, panicle lax, racemes spicate, seeds acutely keeled. I. Very common, K., D. I have only seen a plant or two at Sandpits ! III. i. Campamento ! First Pine Wood, &c. ! ii. Algeciras Station ! Palmones Playazo ! iii. Salt Pans ! C. 02ml if oliuvi Qchv&d. Similar places; rather frequent ; 6-10. Leaves shorter, broader, even upper not narrowed at base, seeds with a thick obtuse keel. It is more general than my stations show. I. Eeclamation Eoad ! III. i. Cachon I C. murale L. Similar places ; common ; 1-12. Shorter, denser, much more leafy, leaves often shining, strongly sinuate- lobed. I.! III.! G. Vulvaria L. Eoadsides and waste places ; rare ; 6-10. Small, prostrate, with a very foetid odour. III. i. Beyond Cam- pamento 1 ii. Algeciras Station ! [C. Bomts-Henricus L., communicated to Kelaarfc, was doubt- less an error.] Beta maritima L. Cultivated fields, roadsides, &c. ; common ; 4-6. I.! IL! III.! fVar. erecta G. & G. A strict erect form, with leaves mostly cordate, appears to correspond with this variety. I. Europa Glacis ! [Phytolacce^.] [Phytolacca dioica L. is often planted, and P. decandra L. is half naturahzed in old gardens.] P0LYGONACE.E. Emex S2niwsa Csbimgd. Waste places; common; 2-4. I.! II.! III. i. and ii. ! Bumex crispus L. Ditches and damp places; rather frequent ; 4-6. Panicle close and dense, leaves narrow, undulate, perianth segments I'ather large, entire. I. About the forts, D. III. ! B. conglomeratus Murr. Similar or drier places ; rather common ; 5-6. Panicle laxly branching, leaves narrow, flat, perianth segments small, entire. II. ! III. ! A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 91 j;B.Friesii G. &, G. Damp grassy places ; rare? 5-6. Panicle much as last, perianth segments spinose-dentate. III. ii. S. de Luna in Los Barrios district, Nilss. B. pidcher L. Dry gravelly and sandy ground ; common ; 4-6. Panicle divaricately branched, leaves panduriform, perianth segments spinose-dentate. I. Near Farringdon's ! Governor's Cottage! II.! III.! B. hucephalopJiorus L. Chiefly cultivated fields ; very abun- dant; 4-5. I.! III.! fVar.j;e?-e?i«a«s \Vk. is aperennant form. I. II. or III. Mobile sand at Gibraltar, and grassy places in the Bay, Wk. II. D. III. i. Sand desert, If., D. iB. thyrsoides Desf. {B. intermedins Guss., non DC). Sandy and waste places ; rather frequent ; 4-5. Panicle dense, leaves lanceolate, fruit perianth large, entire. I. ! III. i. Carteia ! Alcadeza ! ii. ! iii. Salt Pans ! [B. intermedins DC. non Guss. has quite linear leaves, and a lax panicle. A specimen from San Eoque so labelled by Ball is B. thyrsoides.] B. tingitanm L. In deep sand ; rather local ; 4-5. Leaves triangular hastate, often crenately incised. I. Boiss. II. D. III. i. Linea ! Puente Mayorga I ii. Palmones Playazo ! iii. Palmones Village ! jB. scutatus L. Rocky and stony places ; rather rare ; 4-5. I. Slopes over Europa ! Rocks below Europa Point ! Devil's Gap! III. ii. Railway near Algeciras I fVar. glaucus DC. non Boiss. is a very glaucous form, but not otherwise differing from type, with which it grows on the Rock ! Var. induratus Ball is very glaucous, the branchlets becoming indurated and spinescent, and outer perianth segments more reflexed. I. Rocks at Europa Point, iiT., Z).? I think it probable that var. glaucus DC. has been mistaken for it. Polygonum equisetlfonne Sibth Sz Sm. Sandy ground near sea; rare; 6-10. 3-4 ft., stem thick and woody, branches erect and virgate, or sometimes drooping, inflorescence leafless. III. i. Almendral ! Guadarranque Ferry ! ii. Palmones Playazo ! iii. Palmones Village ! Guadacorte Marshes ! P. maritimum L. Deep sand near the sea ; occasional, formerly frequent; 4-11. I. North Front Sentry Fence! Catalan Bay! II. K., D. III. i. Linea, K., D. ii. Algeciras, Nee. iii. Pal- mones, -ST., D. P. aviculare L. Waste and cultivated ground; common ; 4-11. Varies less than in Britain. I think var. agrestinum Jord. is common, and var. arenastrum Jord. less so. I. North Front ! Reclamation Road ! III. ! P. Boherti Lois. (P. Bali Bab., P. aviculare var. vegetum Ledeb. ?). Waste and sandy ground ; frequent; 4-11. I retain this name to cover a species otherwise resembling P. Baii, but much stouter, and obviously perennial, with a thick woody stem. It is sometimes so large as to resemble P. equisetiforme, though usually 92 A FLORA OF GIBBALTAR AND THE NEIGHBOURHOOD prostrate, and with quite different inflorescence, but I fear I may have sometimes confounded the two. II. Neutral Ground! III. i. Eoadside just beyond Campamento, and many other places ! P. H^jclropiper L. Marshy places; locally frequent; 7-10. Eacemes lax, rather nodding, taste biting. III. i. Cork Wood Sotos ! ii. By Miel from source almost to Algeciras ! \P. serrulatum Lag. By streams ; rather frequent ; 1-12. Eacemes erect, compact, flowers pink. III. i. By Lajo and its branches ! Cork Wood Sotos ! ii. ! P. Persicaria L. Vegetable gardens ; occasional ; 5-7. Eacemes dense, rather short, many fruits trigonous. III. i. Near First Venta ! ii. Palmones Playazo ! Algeciras, Bev. P. lapatlii folium L. Similar places; rare; 5-7. Racemes larger and denser, fruits lenticular. Ill, i. Almoraima ! [P. Convolvulus Li. Cultivated ground? casual; 6-9. I. Lag. Found by no other collector.] Thymelace^. Daphne Gnidmm h. Bushy places; common; 6-10. I.! III.! D. Laureola L. var. latifolia Coss. Woods; rare; 2-5. III. ii. Summit of Waterfall Valley ! S. de Luna in Los Barrios district, Larjuna. This may be the same station. Thymelcsa canescens Endl. Heathy hill slopes ; locally com- mon ; 1-4. An erect, branched shrublet, leaves densely white- tomentose, flowers in small clusters, perianth lobes much shorter than tube. I.? Brouss.\ III. i. From Pedrera to Majarambout Crags ! Near First Pine Wood I T. villosa Endl. Similar places, and in mountains ; rather frequent ; 5-6. Diffuse, greyish hirsute, flowers solitary, axillary. I. K. II. Abundant, Boiss., Juss., K. Not there now. III. i. With last ! ii. Mountains to summits ! T.hirsuta Endl. Similar places; rare; 10-4. A tall shrublet, with densely imbricate white-woolly leaves, corolla lobes as long as tube. II. D., not there now. III. i. Linea and S. Carbonera, K., D. Laurine^. Laurus nobilis L. Mountain valleys ; locally abundant? 3-4. III. ii. S. de Palma, Wk. ! I have not seen this in flower or fruit, but think I have seen leaves in several places. Santalace^. Osyris lanceolata Hochst. (0. quadripartita Salzm.). Eocks and bushy slopes ; common on Eock, rare elsewhere ; 5-6. 1. 1 III. i. Cork Wood and Majarambout Crags! 0. alba L. Similar places; rare or error? 3-5. Leaves much smaller, male cymes forming a leafless raceme. I. Tournef. ? Leman's specimens from Gibraltar are certainly 0. lanceolata and not this (see K. Fl. p. 174). III. i. Mountain region of San Eoque, K., D.? Tliesmm hwnile Vahl. Stony rough ground ; rare ; 2-4. I. Above Levant Battery ! a flora op gibraltar and the neighbourhood 93 Aristolochiace^. Aristolochia hcetica L. {A. glauca Desf.). Eough bushy places; abundant on Eock, rather rare in Spain ; 11-5. I. ! III. i. Cachon ! Alcadeza Crags ! First Pine Wood ! ii. Carnero Hills ! Alge- ciras, Rev. A. longa L. Woods and open grassy places ; frequent ; 2-5. I. K. III. i. Upper part of Campo Common ! Carteian Hills, &c. ! ii. and iii. ! [A. rotunda L. has a globose tuber, and a closed sinus to the leaves. I. K. Never confirmed.] Euphorbiace.e. '■'■'■Ricinus communis L. var. africanus Mill. Waste places near buildings ; occasional. The var. is a shrub. I never saw it annual. I. Keclamation Eoad! Sandpits! Lower Lines! III. i. San Eoque ! Puente Mayorga ! Almoraima ! ii. About Algeciras I Mercurialis elliptica Lamk. Heathy ground ; locally frequent ; 3-5. Shrubby, erect and stiff, quite glabrous, leaves subcoriaceous, rather finely crenate. III. i. Second Pine Wood to Long Stables ! f If. BevercJioni Eouy. Bushy places and in woods ; locally common ; 1-6. Perennial, straggling, flaccid, leaves deeply incised. III. ii. All over mountains ! Near Cachon Farm ! Lobo Valley ! M. annua Lt. Waste places; abundant; 1-12. Very variable in size, and in colour and shape of leaves. The complete female appears rare. I. ! III. ! Var. amhigua Duby is as common as type. I. ! III. ! Crozoijliora tinctoria A. Juss. Cultivated ground ; rare ; 6-8. Annual, with some look of a Heliotropium, but very different in- florescence. III. i. Opposite Francia's Farm I By First Pine Wood ! iii. Guadacorte ! Euphorbia Peplis L. Sands near sea ; rare ; 5-9. Leaves obliquely truncate at base, seeds large, smooth. II. Hurst ! III. i. Sand dunes, Brouss., Duf., dx. iii. Palmones, D. ■\E. gibraltarica W.-Dod in Journ. Bot. 1914, 13. Sandy or gravelly places ; rare ; 5-9. Like last, but leaves larger, blotched with reddish black, smaller capsules, and small blackish, 3-keeled, foveolate seeds. III. i. Eailway at Second Venta ! E. Chamasyce L. fvar. canescens Boiss. Sandy or gravelly places ; locally frequent ? 5-9. Much smaller, seeds tetragonous, reticulate rugose. I. In small patches all over the Neutral Ground, and cannot fail to attract attention, K. In Fl. p. 70, Kelaart says the station is the North Front. I have not seen it in either, but it is easily overlooked. III. i. Sand Desert, D. Eailway near Second Venta ! E. adenocarpa Guss. Cultivated fields ; occasional ; 4-5. Tall annual, glands entire, capsule and seeds quite smooth. III. i. Between San Eoque and S. Lorca ! ii. Near Cortijo Trinidad ! E. verrucosa Lamk. Damp woods and meadows ; rare ; 4-6. Perennial, 6-12 in., glands entire, capsule with cylindrical warts. 94 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD seeds smooth, leaves finely serrulate, glabrous. III. i. San Koque, Fritze. \_E. flavicoma DC. Heathy places ; rare ; 4-6. Like last but shorter, capsule warts hemispherical, leaves closely serrate. I. Dasoi'}] \E. rupicola Boiss. Bushy rocky slopes ; locally abundant ; 12-6. Shrubby, much branched, 2-4 ft. or more, involucral leaves bright yellow, capsule densely warted. I. Chiefly Mediter- ranean Steps and Middle Hill ! A form with densely pubescent leaves occurs at the former place ! III. i. Alcadeza Crags, a much less branched form ! ii. Mountains ! ? E. puhescens Vahl. Marshes ; common ; 3-8. Stout, little branched ; very villous, glands entire. III. ! ■\E. iMrococca Brot. Dry fields ; common ; 2-5. Small, dark green, annual, glands entire, capsule with wavy wings. I. Near Michael's Cave ! III. ! E. Helioscopia L. Eoadsides and waste places; common; 1-5. I.! II.! III.! E. exigua L. Fields and grassy places; abundant; 2-5. Leaves acute. III. ! Var. retusa L. Similar places ; more abundant. Leaves shorter, retuse. Looks distinct but intermediates occur. III. ! Var. tricuspidata Koch. Leaves dilated and tricuspidate at apex. III. i. Between Gibraltar and San Koque, D. Malaga Gardens ! ? ii. Beyond Carnero Point ! ? E. Peplus L. Eoadsides and waste places ; very common ; 1-5. Fovese of seeds four on back and three on sides. I. ! III. ! E. peploidcs Gouan. Similar places; common? 2-4. Like last but usually smaller, seeds with three foveae on back, two on sides. I.! II.! III. i. San Koque, D. ii. Algeciras, i?a//. E. falcata L. Similar places ; rare '? 4-6. Kather like E. Peplus but leaves broader, acute, capsule deeply sulcate and keeled, seeds transversely sulcate. III. i. Beyond Linea Cemetery ! Cruz del Padre Ventura ! Pinar de los Bigotes ! iii. Kailside near Guadacorte ! E. medicaginea Boiss. Sandy fields and gravelly slopes ; abundant ; 3-5. Annual, with bright yellow involucral leaves, glands bicornate, seeds black with white reticulate ribs. I. About Willis's ! III. ! Often forming large yellow masses in fields ! E. segetalis L. Cultivated fields and woods ; rather rare ? 3-6. Annual, dull glaucous green, capsule granulated on back, seeds ashy grey, irregularly foveolate. [I. Wrongly attributed to Von Martins by Kelaart.] III. ii. Slopes above Waterfall Valley ! I think I have seen this elsewhere, but it closely resembles E. terracina externally, and I may have taken it for that. E. portlandica L. Kough bushy places and woods ; locally common ; 3-5. Perennial, with closely imbricated dark green leaves, capsules granulated and seeds foveolate. I. From Europa to above Mediterranean Koad! II. K., D. III. i. Sand dunes and Spanish racecourse. K., D. ii. Mountains over Pelayo ! I Var. intermedia Porta & Rigo. A tall very stragghng form. A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 95 III. i. Cork Woods ! Majarambout Woods ! I do not know whether the name has been pubhshed. E. bcBtica Boiss. {E. trinervia Boiss.). Sandy places and among bushes ; locally frequent ; 3-6. Perennial, laxly branched, leaves narrow, capsule longly stipitate, smooth, seeds foveolate. I. Above Catalan Bay ! Sea sand near Inundation, K. III. i. Bond's Farm ! Near Spanish racecourse, Boiss., Kel. Alcadeza Plain ! Cork Woods ! E. serrata L. Sandy and bushy ground ; rare ; 3-6. Leaves strongly serrate, the involucral bright golden yellow. I. Lem. ! III. i. Bond's Farm ! Foot of S. Carbonera and of San Eoque, K., D. Pinar de los Bigotes ! ii. Foot of S. de Palma, K., D. iii. Railway near Guadacorte ! E. terracina L. {E. provincialis Willd.). Sandy places ; frequent ; 2-5. Perennial, many stemmed, capsule and seeds smooth. I. North Front ! Reclamation Road ! Sandpits ! Victoria Battery ! II.! III. i.! ii. Algeciras, -K"., Z). Var. latifolia Boiss. has broad leaves. II. K., D. fVar. angustifoUa Lange has narrow leaves. II. K., D. III. i. Lajo below First Pine Wood ! t Var. retusa Boiss., leaves obovate, deeply obcordate. II. K., D. [E. Esula L. is wrongly attributed to Von Martins from the Rock by Kelaart.] E. Paralias L. Sea sand ; locally rather frequent ; 5-8. Perennial with broad, coriaceous, adpressed leaves. I. Ayala. II.! III. i. Linea, Jf. First River! ii. Palmones Playazo! iii. Palmones Sands ! E. Characias L. Wooded and rocky places ; locally frequent ; 2-4. Tall, very stout, unbranched, often 3-4 ft. high, inflorescence racemose, glands purple, entire. I. About Signal Station Road ! Always quite glabrous. III. i. Cork Woods about Almoraima ! Always pubescent. Calliteichace^. CalUtriche verna Kuetz. Ditches or streams; rare? 5-10. Leaves mostly obovate, styles erect, deciduous, fruit with acute keel. An ambiguous species, probably included in next. III. i. Near San Roque, Wk. ? C. stagnalis Scop. Similar places ; frequent ; 4-5. Leaves as last, styles diverging, persistent, fruit winged. III. ii. In the mountains ! Palmones Playazo ! \C. Jiaimdata Kuetz. Similar places; rare? 4-5. Leaves, at least lower, linear, keel of fruit winged. III. i. Campo Common! ii. Near Palmones Pinar ! Urticacejb. Theligonum Cynocramhe L. Rocks, walls and banks ; locally common ; 2-4. Stem and bright green leaves subsucculent. I. ! III. i. Cachon ! S. Carbonera, D. Parietaria mauretanica Desr. var. latifolia Lge. Rocky and bushy places ; local ; 1-5. Leaves very broad, subcordate. I. North slopes, both sides ! P. diffusa Mert. & Koch. Similar places ; abundant on Rock, 96 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD less SO in Spain ; 1-12. Very variable. I. ! III. i. Seen but no stations noted ! ii. Palmones Piayazo ! About Algeciras ! Carnero Hills ! Var. fallax G. & G. has subsimple erect stems. I. ! I think frequent. Mr. Druce thinks my 1579, from Eock Gun Catchment, may be P. erecta M. & K., a distinct species like y&r. fallax, but with perianth not elongating after flowering, but the specimen is too young for determination. P. lusitanica L. Walls and rocks ; rare ? 3-5. Like a diminutive diffuse P. diffusa, but leaves darker green, rounder, and glomerules subsolitary. I. Main road near St. Bernard's ! Below Devil's Gap ! Urtica tirens L. Waste and cultivated ground, chiefly near buildings ; very common ; 11-5. I. ! II. ! III. ! U. memhranacea Poir. Similar places ; equally common ; 11-5. Usually dioecious, though described as monoecious. The female appears about six weeks after the male. I. ! II. ! III. ! [U. dioica L. is reported as common on the Kock by Kelaart and Dautez, but by no other collector.] Celtis australis L. Hedges and rough slopes, often planted ; 4-5. I.Levant! Signal Station Eoad, &c. ! III. i. La jo Valley! ■■'•Ficus Carica L. Kocks and stony places ; completely natura- lised in very many places ; 4-6. I. ! III. ! Reaches high up the mountains ! [Ubnus ca^npestris Sm. is probably always planted, as about Algeciras, El Cobre, Lajo Valley, &c. ! I believe this species, but I have not seen fruit.] [Morus nigra Willd. is only planted.] CUPULIPER^. Quercus lusitanica Webb. Woods ; common ; 3-4. Leaves very large, more or less deciduous. Var. faginea Lge., the type, has dentate or subentire leaves, and appears rare. III. ii. S. de Palma, Boiss., Wk., dc. Var. bcetica Webb has coarsely sinuate crenate leaves, and is common. III. i. Cork Woods ! ii. Mountains I Q. humilis Lamk. Woods and heaths ; very common ; 3-4. Leaves incanous beneath, hardly spinose. I think it seldom grows with Q. coccifera. III. i. and ii. ! Q. Suber L. Woods ; abundant ; 4-5. [I. Introduced.] III. i. and ii. ! Q. Suhcr X Ilex Lag. With the parents ; rare ; 4-5. III. i. Near San Eoque, D. Q. Ilex L. Woods ; rare ? formerly frequent ? '4-6. I. In- accessible places on west slopes, K. Near Breakneck Battery, two trees I Above Catalan Bay, one tree ! III. i. Cork Woods ! Q. coccifera L. Heaths and woods ; very common ; 3-5. Leaves shining, quite glabrous both sides, often very spiny. I. Under the shade of larger trees, K. Are these not root shoots of Q. Ilex? I have not seen it. III. i. and ii,! In deep sand at Palmones Piayazo ! a flora of gibraltar and the neighbourhood 97 Amentifer^. Alnus glutinosa Gaertn. var. doiticulata Eegel. By streams, chiefly in woods ; locally plentiful ; 1-3. III. i. Cork Woods ! ii. By many streams from the mountains above Algeciras ! iii. Salt Pans ! Salix alba L. By water holes and streams, doubtless often planted, but native in many places ; 3-4. III. i. About San Roque ! Cork Woods ! ii. ! iii. About Salt Pans ! fS. cinerea L. Similar places; frequent, commonly planted by water holes; 1-3. III. i. By the Lajo! Cork Wood Sotos! ii. Waterfall Valley ! Carnero Hills ! iii. Salt Pans ! S. 2^&dicellata Desf. Similar places ; occasional or frequent ; 1-3. Catkins produced on pedunculate leafy shoots, leaves like last. III. i. By Lajo at First Pine Wood! ii. Waterfall Valley! [S. hahylonica L. is occasionally planted, as by river above San Roque Station !] Populus alba L. Woods and by mountain streams, often planted, but here and there native ; 1-3. [I. Planted.] III. i. Near First Pine Wood ! Almendral ! Almoraima ! ii. Valley over Frayle Bay ! iii. Guadacorte ! [P.pijramidalis Rozier, the Lombardy Poplar, is often planted.] CYTINACE.aE. Cytinns Hypocistis L. On roots of Cistacea ; occasional ; 3-5. III. i. Summit of Chair! Cork Wood Crags ! Pine Wood Plains ! ii. Mountains south of Waterfall Valley ! Conifers. \Pinus halepensis Mill. Woods; rare; 4-5. Cones rather long, shining, deflexed. III. i. Between San Roque and Castellar, D. [P. sylvestris L. and P. Pinaster Ait, are planted. They both have small cones, P. Pinaster having longer leaves and larger cones, scales with acute ridges.] P. Pinea L. Woods ; frequent ; 4-7. Cones large and ovoid or subglobose, scales with obtuse ridges. III. Forms the Pine Woods in many places, but also planted ! Juniperus Oxy cedr us 1j. Rocky ravines ; rare; 3-7. I. West slopes, Schott ? III. i. Hills above S. Carbonera, D. Gnetace^. Ephedra fragilis Desf. Rocky slopes and sandy places ; frequent on Rock, occasional in Spain ; 4-6. Dr. Stapf says all he has seen from the Rock is the type, i.e. var. Desfontainei Stapf. It varies much in habit. I. ! Especially about Levant ! III. i. Puente Mayorga ! ii. Palmones Playazo! Rocks at Alge- ciras, Wi7ikl. 1 [E. altissima Desf. {E. gibraltarica Boiss. ?). Dr. Stapf says all he has seen from here so named is E. fragilis. The true E. altissima is a very different species.] Journal of Botany, July, 1914. [Supplement.] ]i 98 a flora of gibraltar and the neighbourhood Lemnace^. Lemna gibha L. Water holes; occasional; 4-6. I. Near Inundation, de Coincy. III. i. Linea ! ii. Palmones Playazo ! iii. Palmones Sands ! L. minor L. Similar places ; locally frequent ; 4-6. III. i. Lajo below Almendral ! ii. and iii. With the last ! Naiadaces:. iZannichelUa macrostemon J. G&y. Pools; rare; 4-6. III. i. Near San Eoque, Nilss. ii. ! ? My 1112, a young state, from a roadside pool towards Palmones Pinar, may be this or Z . ixtlustris L., a frequent species in the province. \Potamogeton amoricanus Cham. & Schl. (P. fluitans Roth). Streams ; locally common ; 6-8. III. i. In the Lajo and its tributaries ! ii. In mountain swamps ! fP. pusilhts L. var. elongatus Ar. Benn. Streams ; rare ; 5-6. III. i. In the Lajo and its tributaries ! [Rupina rostellata Koch is reported from the Inundation and Neutral Ground by Kelaart and Dautez, and is attributed to Boissier by Debeaux, but Boissier's specimen is correctly labelled B. maritima !] B. maritima L. [B. spiralis Dum.). In salt water; locally abundant; 8-10. Peduncles spirally coiled. I. The Inundation is full of it ! Near the convict station, A'. ; now built on. III. ii. and iii. Salt marshes at San Roque and Algeciras, Nilss. Zostera marina L. In the sea or tidal rivers ; frequent ? 6-7. I. Western shores, K. II. and III. All round the Bay, K., D. ^ Posidonia Caulini Koenig. In the sea ; rare? 4-8. With a large tuft of dead leaf sheaths at base, flowers in longly peduncled spathes. I. Near the North Front Guard, Kel., Boiss. III. ii. Algeciras, D. Alismace^. ■\Alisma ranunculoides L. Pools, ditches and streams; very common; 4-6. II.! III.! A. Plantago L. Similar places; much less common; 4-6. The type has leaves broadly rounded or cordate at base, and is rare. III. i. Sand desert below Pedrera, D. Var. lanceolata G. & G. with leaves narrowed below, is the usual form. III. i. Lajo Valley! Carteian Hills! Campamento Common! ii. Algeciras, i?ey. Palmones Playazo ! iii. Salt Pans! Juncaginace^. Triglochin maritimuvi L. Salt marshes; rare'; 4-5. Spikes dense, fruit of six carpels. III. i. Sand dunes below S. Car- bonera, D. T. Barrelieri Lois. Fresh water marshes ; occasional ; 1-3. Spikes very lax, fruit of three carpels. III. ii. Palmones Playazo ! iii. Guadarranque Marshes ! [T. palustre L. is recorded from near Gibraltar by Schott {ex Colm.) in addition to T, Barrelieri, but is unconfirmed.] a flora op gibraltar and the neighbourhood 99 Aroide.^. Arisarum vulgare Kunth. Dry open and bushy places ; very abundant; 11-1. I.! III.! Arum italicum Mill. Shady banks; rather frequent, and quite native ; 3-4. I. Sandpits ! Above Willis's ! Above Engineer Road! III. i. Malaga Gardens! ii. and iii. ! [A. mamdatum L. is recorded from Algeciras by Clemente, but is unconfirmed.] [A. Dracuncidus L. is subspontaneous here and there, but I have not seen it.] '''Colocasia antiquonim Schott. Shady ravines ; quite natura- lised in one spot ; 6-7. III. ii. Valley below Waterfall ! Typhace^. Typha latifolia L. Pools and streams ; rather rare ; 6-7. My gathering has rather narrow leaves, 3-3| lines wide, and male spike about J in. above female. It may be a hybrid. III. i. Near Almoraima I ii. In the Miel ! iii. Guadacorte Marshes ! Salt Pans ! T. angustifolia L. Similar places; locally common; 6-7. More frequent than my records show. III. i. In the Lajo ! ii. Carnero Hills ! iii. Guadacorte Marshes ! Salt Pans ! Sparganium ramosum Huds. In streams ; locally common ; 4-7. III. i. In the Lajo ! Cork Wood Sotos ! ii. In the Miel ! Marsh below El Cobre ! Towards El Saladillo ! Palmes. Ghamarops humilis L. Bushy hills ; very common ; very rarely in woods or heathy places ; 4-5. I. ! Often with a trunk 4-6 ft. high. III. I Orchidace^. Serapias cordigera L. Woods, rarely open ground ; rather locally frequent ; 4-5. I. West slopes, iT., ScAo^^. III. i. Queen of Spain's Chair! S. Lorca ! Cork Woods! ii. M. de la Torre ! Palm ones Pinar ! Waterfall Valley, especially at El Cobre I My 1895, from marshy open spots near railway beyond upper Miel Bridge, is a peculiar form in an unusual situation, with small flowers, and a rather narrow lip, almost intermediate between type and S. Lingua, but Mr. Rolfe thinks it belongs here. S. occidtata J. Gay. Damp grassy places ; rare? 4-5. Flowers smaller than next, hidden in bracts except lip, which is short, narrow, and reflexed, the basal ridges pale and distant. III. ii. Near i\lgeciras, Bcv., Frere. Carnero Point ! S. Lingua L. Similar places ; very common ; 4-5. Lower perianth lobes more exposed, lip nearly twice as long as last, and broader, basal ridges dark and close, so that they are described as one. I. ? West slopes, Schott. II. ! III. ! Orchis papilionacea L. Bushy ground; very rare; 4-5. Lip large, reddish purple. III. i. Magazine Hill, a single plant ! 0. Morio var. picta Reichb. Woods and bushy places ; rare ; 3-5. Spur truncate, almost as long as ovary. III. i. Between /; 2 100 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD Almoraima cancl Long Stables! North slopes Alcadeza Crags! About San Eoque, Wk. 0. coriophora L. var. Polliniana Keichb. f. [0. fragmns Poll). Hedges; very rare; 4-5. Flowers rather small, livid purple, in a dense spike. III. i. A plant or two by railside beyond Almoraima ! 0. lactea Poir. (0. variegata All. var. acuminata Boiss.). Eough slopes; rather rare; 2-3. I. Below Signal Station! Spy Glass ! i. Grassy places in woods near San Eoque, Wk. 0. longicruris Link. Similar places ; very rare ; 3-4. Allied to 0. Simia Lamk. I. '? West slopes, Schott. 0. laxiflora Lamk. var. longibracteata "Wk. Marshy ground ; very common; 3-5. The var. has a dense spike, with bracts longer than ovary, the lowest foliaceous, and is alone recorded, but I think type is found frequently. III. ii. and iii. ! No note from i., but I believe it occurs. to. pakistris Jacq. Woods; very local ; 4-6. Flowers much paler, lower lip 3-lobed, the central largest and longest, usually divaricately lobed. III. ii. Upper Waterfall Valley ! 0. longibracteata 1M\. Woods; rare; 3-4. Leaves very large, sepals connivent, flowers large, lip rose-purple, spotted, 4-lobed, the two central divaricate. III. i. S. Carbonera, Frere. ii. x\lge- ciras, Frere. fO. cordata Willd. Eough shady slopes; locally rather fre- quent ; 2-4. Flowers green, leaves broadly ovate, cordate. I. Upper Eock from Lower Union Gallery to Mediterranean Steps ! Eocks above Catalan Bay ! III. i. Cork Wood Crags, rare ! ii. Waterfall, rare ! Ophrys aranifera Huds. var. atrata Eeichb. Eocky slopes ; rare; 3-4. Lip longer than sepals, subtrilobed, blackish purple, basal bosses conical, sahent. I. South and west slopes, Frere, Schott, D. A specimen from the Eock by Lemann, labelled 0. aranifera, is 0. fusca ! 0. tethredinifera Willd. Grassy and bushy slopes; locally rather frequent ; 2-4. Eather like next but shorter, with a shorter spike. I. Communicated, K. III. i. Queen of Spain's Chair ! Carteian Hills ! North of San Eoque I ii. North of Algeciras ! El Cobre ! Carnero Point ! 0. apifera Huds. Watercourses and marshes ; rather common ; 4-5. Occasionally almost white. I. About Willis's ! Glacis, a single specimen ! South slopes, D., Bev. III. i. Carteian Hills ! Campamento Common ! Long Stables ! ii. ! iii. Guadacorte ! Salt Pans ! 0. Scolopax Cav. Dry bushy places ; rather rare ; 4-5. Eather like last, but flowers much smaller, usually white, rarely pale rose. III. i. Near San Eoque Station ! S. Carbonera, D. By Arroyo Viejo near Patxot's farm! Alcadeza Crags! ii. About El Cobre ! Hills west of Algeciras ! O.bovihyliflor a hink. Grassy places ; frequent ; 2-4. Flowers greenish, lip dark, wdth very acute bosses. I. One or two specimens near Signal Station ! II. ! III. i. Path to First Pine Wood, and on neighbouring hills ! ii. Common, to Carnero Point ! A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 101 0. Speculum Link. Similar places ; rave ; 3-4. Lip with a large dark blue shining patch, and a papillose fringe. I. Two specimens below Signal Station ! III. i. Path to First Pine Wood ! Cork Wood Crags ! ii. S. de Palma, Wk. 0. ftisca Link. Eough and grassy slopes ; locally rather frequent ; 1-3. Lip narrow, dark, with very narrow yellowish edges. I. West slopes! III. i. North of San Eoque, occasional! Carteian Hills ! ii. Algeciras, Frere. 0. lutea Cass. Similar places ; rather frequent on Eock, rare in Spain ; 2-4. Lip broad, with a broad yellow margin. I. North- west slopes to Signal Station ! Debris slopes over North Front I III. i. Queen of Spain's Chair, common, K. I have never seen it there. Malaga Gardens ! ii. Near Sandy Bay ! Hills near Algeciras K. iii. Guadacorte ! ■\Neotinea intacta Eeichb. [Aceras derisiflora Boiss.). Woods ; rare ; 4-5. Spikes dense, unilateral, flowers small, pinkish white. III. i. Queen of Spain's Chair, rare, K. ii. Between Waterfall Valley and Pelayo ! Spiranthes autuvinalis Eich. Eocky places ; rare ; 10-11. Leaves broadly oval, after flowers, flowers yellowish white. I. From Willis's to Signal Station ! \S. cestivalis Eich. Damp grassy places ; very rare ; 5-6. Leaves lanceolate, contemporary with white flowers. III. i. East slopes Queen of Spain's Chair ! ii. Beyond Waterfall ! Epipactis atroruhens Schultz. Woods ; rather rare ; 4-6. Flowers rather deep red, lip with crisped plicate bosses. III. i. First Pine Wood ! ii. Mountains ! CephalantheraensifoliaHich. Woods; rare; 3-4. III. [i. Boca de Leon, Hurst !] , beyond our limits, ii. Slopes south of Waterfall ! ILimodoruviabortivtimSw. Woods; rather rare ; 4-6. Whole plant lavender and white. III. i. Near Long Stables! Majaram- bout Woods ! ii. Waterfall Valley ! Iridace^. Gladiolus segetum Gawl. Cornfields ; rather rare ? 4-5. Flowers large, distichous, middle perianth segment twice as broad as lateral, seeds not winged. Confounded with next, and certainly not our common species. I.? K. III. i. Fields by Lajo ! Pinar de los Bigotes ! ii. Near railway beyond Algeciras ! G. communis L. Hills, bushy places and fields ; very common ; 4-6. Flowers much smaller, unilateral, middle segment not much broader than lateral, seeds broadly winged. I. ! III. ! Eeaches high up the mountains ! Iris Sisyrinchium L. Dry fields and roadsides ; common ; 2-4. I.! II.! III.! I. Fontanesii G. & G. Grassy places ; very rare ; 4-5. Eesembles /. Xiphium L., with large blue flowers. III. ii. Near upper aqueduct ! Eailside near San Bernabe ! One or two plants in each place. 17. filifolia Boiss. Eough heathy and bushy slopes ; frequent and locally common ; 4-5. I. Upper Eock from north to south I 102 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD III. i. Queen of Spain's Chair, Porta d: Rigo ! South slopes of San Eoque, D. Alcacleza to Majarambout! ii. Waterfall Valley, plentiful ! \I. juncea Poir. (J. lusitanica Ker). Similar places; rare; 5. Like last, but flowers yellow, tube much shorter. III. ii. Mountains near Algeciras, Bev. I. Pseudacorus L. Swamps and watercourses; locally com- mon ; 2-4. III. ! Common in Cork Woods ! I. fostidissima L. Woods ; rather rare ; 5-6. III. i. Almo- raima ! ii. Mountains ! Carnero Hills ! I. scorpioides Desf. (I. alata Poir.). Grassy sandy places ; rare ; 1-2. III. i. Slopes south of San Eoque, D. BotmUea Clusiana Lge. Sandy and grassy places ; locally abundant ; 1-3. Flowers very large, with a bright yellow base. I. Chiefly Europa Flats, Windmill Hill, and North Front! Occasionally on upper west slopes ! II. ! III. i. Near Spanish racecourse, Chis., D. B. Bulhocodium Ker. Similar places; rare? 1-3. Flowers much smaller, upper spathe membranous, stigmas bipartite, often longer than stamens. [I. Kelaart's note obviously refers to last.] III. i. San Eoque, D. S. de Palma, Bev. Often confounded with B, ramiflora, but it is common in the province. [B. lyurimrascens Ten., recorded by Willkomm and Dautez from San Eoque, was founded upon a garden specimen, allied to the South African B. rosea Eckl.] B. ramiflora Ten. Similar places ; very common ; 1-3. Smaller than B. Bulhocodium, upper spathe herbaceous, stigmas bilobed, usually shorter than anthers. II. ! III. ! B. gaditana Beg. {B. Linaresii Pari. var. gaditana Kunze). Similar places ; occasional ? 1-3. Confounded by Willkomm with B. ramiflora, from which it differs chiefly in its much larger flowers, 2-4 times as long, and deeper in colour. III. i. Sand Desert, D., teste Beguinot. This may be the plant recorded by Xy-Auiez B.'&B.Columnce. Bond's Farm!? ii. Hills near the Miel ! ? Specimens from the last two stations (my 148 and 165) probably belong here, but the upper spathes are almost wholly membranous, instead of half herbaceous. They cannot be referred to any of our other species. B.iMrviflora Britten in Journ. Bot. 1914, 46 {B. ColumncB auct.). Similar places ; occasional ? 1-3. Like B. ramiflora, but flowers very small, pale, seeds dull, angular. II. D. ? III. i. Sand Desert, D. ? It is probable that Dautez has confounded the two. Crocus Salzmanni J. Gay. Bushy places; rare; 10-11. Difiers from C. serotinus Salisb. in orange, not white filaments, and yellowish, not white throat, and broader leaves. Discovered by Maw (Monog. Gen. Croc. p. 103). I. Windmill Hill ! Am ARYLLID ACE^ . Leucojuvi trichophylUim Brot. Sandy or light soil; very rare; 1-3. III. i. [In great profusion on Spanish racecourse, K. Kelaart obviously refers to next.] In 1883 I found this in small A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD ' 103 quantity by the path to Eocaclillo, which is now all cultivation and I think it is exterminated ! L. mttumnale L. Similar places ; locally abundant ; 10-1. [I. A few specimens at Willis's and Ince's, where I planted it in 1883 !] III. i. Punta Mala ! Campo Common ! Carregnoa humilis J. Gay. Similar places ; very rare ; 10. III. i. Carteian Hills near Puente Mayorga ! Near Almoraima, D. Pancratium maritimum L. Sand dunes by sea ; locally abundant; 7-9. I. North Front ! II. ! III. ! Corhularia Bulbocodium Haw. Kocks on tops of mountains ; locally common ; 12-3. III. ii. On all the highest ridges ! Narcissus serotinus L. Gravelly places ; rather frequent ; 10-11. Tall forms with 2-4 flowers are var. major P. L. III. i. Campamento Common to Pindalista ! Alcadeza Plain ! Probably elsewhere, but not seen in fiow^er. \N. viridiflorus Schousb. Damp gravelly and clayey places ; frequent ; 9-10. In 1883 I gathered on Campo Common, with Maw, a hybrid between this species and the last. III. i. Sand Desert, D. Campo Common to San Eoque ! Carteian Hills ! Second Venta ! ii. Palmones Pinar to Sandy Bay, and commonly near Algeciras ! N. niveus Lois. Eough bushy places or marshy ground ; common, locally abundant ; 10-3. On the Kock this grows in dry bushy spots, flowering from October to February ; in Spain it inhabits marshes, and flowers from January to March. It varies greatly in shape of perianth segments. I. ! III.! [N. iwlyanthus Lois., a frequent species in the province, with subterete scape, and pale yellow entire corona, was found on the Eock by Clusius, but never confirmed.] [Agave americana L. is very extensively planted, and has become so well naturalised here and there on the Eock and in Spain as almost to deserve inclusion.] D10SCORACE.E. Tamus communis L. Bushy and wooded places ; rather frequent ; 2-5. I. ! III. i. and ii. ! Common in the mountains ! SMILACE.E. Smilax aspera L. Bushy ground ; frequent ? 9-1. I have never seen the type, which has red fruit (S. rubra Willd. ?), but Debeaux says it is common. The leaves seem to vary too indefi- nitely to afford specific characters. I. K., D. III. K., D. Var. mauritanica G. & G., with black fruit, is common. The fruit in all I have seen is smaller and in larger clusters than in the common species of the Italian Eiviera, though by description it should be larger than in the type. The forms require elucidation. I.! III.! Var. vespertilionis Boiss. has large leaves, much broader than long, deeply cordate. A robust form, climbing tall trees. III. ii. Top of Waterfall Valley ! Seen elsewhere, but no stations noted. Buscus aculcatusLi. Woods; locally frequent ; 2-4. [I. Ala- 104 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD meda (Balestrino !, labelled B. HTjpophyllum). Not native, I think.] III. i. Near San Eoque Station ! Queen of Spain's Chair ! Cork Woods, especially near Almoraima ! ii. Waterfall Valley ! \B. Hypophyllum L. Woods and crevices of rocks ; locally com- mon; 12-4. Floral bracts small, subulate. I.! III. ii. Mountains ! \B. Hypogloss^im L. Similar places; rare ; 2-4. Floral bracts large, foliaceous. I. Winkl. III. ii. S. de Palma, Bev. Asjmragus acutifolius L. Bushy places and woods ; 8-9. Sub- cHmbing, phyllodes dark green, fascicled, 1^-3 lines long, flowers 1-2. I. South and west, A'., D. III. i. About San Roque and S. Carbonera, D. Almoraima Soto ! ? a very long-spined form. A. ap)hyllus L. Bushy places and open fields ; frequent; 3-5. Erect, phyllodes stout, solitary, rarely fascicled, 1 in. or more long. Flowers one or several. III. ! I have noticed this scattered all over the country, but never in flower or fruit, so made no records. Var. sttpularis Baker (^1. liorridus L.) has stouter, very long phyllaries, often 2-3 in. long. III. ii. Algeciras, Nee. A. albus L. Similar places; very common on Rock, occasional in Spain; 9-11. Phyllodes soft, pale green, fascicled, flowers several, fruit coral-red when ripe, not black as described. I. ! III. i. PuenteMayorga! Carteian Hills ! ii. Miel Valley ! Near Palmones Pinar ! Carnero Hills ! iii. Guadacorte! Salt Pans ! [A. officincdis L., reported by Kelaart as all but wild on the Eock, has been recorded by no other collector.] LlLIACE^. Fritillaria hisitanica Wikstr. var. hispanica Baker. Rough slopes ; rather rare ; 3-5. I. West slopes, rare, D., K., Lem. III. i. S. Carbonera, Bev. ii. S. de Palma, Bev., Willk. Will- komm's record is for F. messanensis Raf., which Debeaux says has been confounded with F. hispanica. Var. stenophyUa Baker is hardly distinguishable by its narrower leaves, smaller less tesselated perianth, and shorter style branches. It seems to be our commoner form. III. i. Bonel's Farm ! West slopes Chair ! Sandy places near Cork Woods, Hurst ! ii. Occasional mountain slopes ! iii. Fields near shore at Palmones, Bev. Tulipa australis Link. (T. Celsiana DC). Rough slopes, rare ; 3-4. III. i. South slopes of San Roque, Wk. Near Almoraima, Wk. Common round Queen of Spain's Chair, K., D. I formerly found this commonly above Campo Common, but have recently searched for it in vain. It may have been exterminated by cultivation. Ornithogalum narbonense L. Sandy fields, &c., occasional ; 4-5. Scarcely distinct from next and perhaps confounded with it. Perianth pure white with narrow green keel, filaments longly attenuate, scarcely half perianth. III. i. At foot of San Roque and S. Carbonera, D. to. pyrenaicum L. Similar places; common; 4-5. Taller, perianth with a broad green keel, filaments abruptly acuminate, nearly as long as perianth. II. A plant or two ! III. ! A FLORA OF GIBRALTAR AND THE NEIGHHOURHOOD 105 0. unifolium Gawl. Sandy heaths ; frequent, locally abundant ; 3-5. [I. K. Fl. p. 161.] III. i. and ii. ! Reaches high up mountains. 0. uvibellatum L. var. longihracteatum Wk. Bushy or sandy places, and in woods ; common ; 2-4. I. Levant ! Near Mon- key's Cave ! Europa Flats ! Above Haynes's Cave ! III. ! Reaches top of El Frayle ridge ! Scilla verna Huds. var. major Boiss. (*S. Eamhurei Boiss.). Rough slopes and heathy places ; frequent, but rather local ; 2-4. Taller than type, flowers many, racemose, not subcorymbose, anthers blue. I. Rock Gun ! Below Signal Station ! III. i. Almoraima! Second Pine Wood ! ii. Algeciras, Lc»i. ! Palmones Pinar and Heath ! iii. Guadarranque Marshes ! S. liemispharica Boiss. Rough slopes on Rock, and marshes in Spain ; locally very common ; 3-4. I. Mediterranean Steps ! Governor's Cottage ! North-west slopes ! III. ! Rare in i. ! S. monophylla Link (S. immila Brot.). "Woods ; locally abun- dant ; 1-4. III. i. Queen of Spain's Chair ! Cork Woods ! ii. Mountains ! S. autumnalis L. Sandy places ; abundant ; 9-10. I. ! III. ! Not seen in II., but I have not been there in its flowering time, and no doubt it occurs. Urginea Scilla Steinh. (S. maritima L.). Rough and grassy slopes ; common ; 9-10. I. ! III. ! Rare in woods ! [Endymion canqxinulatus Wk. {Scilla camp. Ait.) is said by Kelaart to form large beds by the road to San Roque, but it is a sylvestral species not likely to occur there, and not confirmed.] Uroj^etalum serotiimim Ker. Sandy fields ; rare on Rock, frequent in Spain ; 3-4. The flowers are always olive-brown or greenish, but often turn quite vermilion on drying. Is this the origin of U. fulvus Rouy, which is described as having fulvous or orange flowers ? I. Mediterranean Steps ! Levant ! II. ! III. ! Muscari comosum Mill. Sandy and grassy fields ; common ; 3-4. I. Below Victoria Battery and above Alameda, Lcm. Europa Flats, Gai)t. Luck. II. ? Abundantly, K. From Kelaart's mention of the races I think North Front is meant, but I have seen it in neither. III. ! Rare in ii. ! Simethis hicolor Kunth. Open woods and bushy hills ; rather locally common ; 2-5. III. i. Queen of Spain's Chair and else- where ! ii. Mountains 1 [Aloe arboresceiis Mill, is only planted.] Asphodelus fistulosus L. Dry sandy soil, rocks, and old walls; locally frequent ; 2-4. I. ! North Front ! II. ! III. i. S. Carbo- nera, D. iii. Salt Pans ! A. microcarpus Viv. Rocky slopes, fields, and woods ; very common ; 1-4. Always spreading-branched, bracts pale, rather narrow, filaments papillose to middle, claws square or twice as long as broad, fruit ovoid, 5 lines long by 4 lines wide, central keel as prominent as angles. I. ! III. I \A. serotinus W.-Dod in Journ. Bot. 1914, p. 13. Woods and 106 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD rocky slopes ; locally common ; 5-6. Habit of last, but rather more slender and glaucous, often taller, bracts broad, fruit small, much narrowed in lower third, 3 lines long by 2| wide. Flowers much later. III. i. From Alcadeza Crags near Second Pine Wood to Boca de Leon ! A. cerasiferus J. Gay. Similar places; locally very common; 2-4. Eaiely branched, spike very dense, bracts broad, pale, filaments smooth except claw, which is four times as long as broad, fruit depressed globose, umbilicate at base, 7 lines long by 8 wide, valves thick, central ridge faint. III. ii. Mountains ! iVar. albus Baker. Similar places; local; 2-4. Doubtfully distinct, occasionally with a long erect branch or two, bracts very dark, fruit globose, not umbilicate, 5 lines long by 6 wide. I. About Levant ! III. ii. Occasional with type ! Allium pcmiciilatum L. var. pallens G. & G. Sandy and stony places ; occasional ; 5-7. Stamens simple, perianth segments truncate, pale, with green or olive band, often turning rosy on drying. The type (not recorded) has rose flowers. I. Old walls, Bev., K., D, III. i. About San Koque and S. Carbonera, D. By Lajo ! Alcadeza Crags! Path to Malaga Gardens! ii. Near Cortijo Trinidad ! iii. Behind Palmones Village I iA. spkcerocei^lialum L. Sandy ground ; rare ; 6-7. Leaves semicylindrical, heads small, compact, flowers deep crimson, ovoid, petals connivent at tips, anthers exserted, middle cusp as long as lateral, and half as long as claw, auxiliary bulbs often some way above main bulb. I. Catalan Bay ! Levant ! III. i. Rail beyond Almoraima ! \A. rubro-vittatum Boiss. & Heldr. Rocky places; rare; 6. Slender, 4-6 in., bulb small, leaves semicylindrical, flowers white with red band. III. ii. S. de Palma, Bev. A. Ampeloprasumh. Stony ground; occasional; 5-6. Stout, bulb large, with many small bulblets, leaves flat, flowers white in type. III. i. Between S. Lorca and Alcadeza Crags! Almoraima Soto, a field full ! ii. Railside beyond Algeciras ! iii. Salt Pans ! tVar. atropurpureum Regel has dingy purplish flowers, locally common. I. ! A. roseicm L. Rough bushy places ; rather frequent on Rock, occasional in Spain. Flowers large, pale pink, few, in a loose umbel. I. ! III. i. Almoraima ! Alcadeza Plains ! ii. Railside near Algeciras ! M. de la Torre ! Carnero Hills ! A. neaiJoUtanum Cyr. Fields and hedges; rare; 3-4. Like last, but perianth more spreading, pure white, stem trigonous. [I., A'. Kelaart was uncertain as to the name.] III. i. San Roque, D. \A. nigrum L. Fields and grassy places ; occasional ; 4-5. Stout, heads dense, flowers dingy pink, perianth stellately spread- ing. III. ! Very common north of San Roque ! \A. Moly L. var. stramineiim W.-Dod {A. stramineum Boiss. & Reut.). Mountain slopes ; locally frequent ; 4-5. Flowers large, A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 107 yellow, in a lax umbel. Leaves narrower and flowers paler than in type. III. ii. Mountains ! A. Cliamcemoly L. Marshy places ; very local ; 11-2. Very dwarf, flowers white, leaves long, flat, spreading. III. ii. By railway near Algeciras ! A. siibvillosum Sa\zm. Sandy ground ; rare; 1-2. Like last, but much taller, leaves densely ciliate. 111., Nee, Navarro, ii. Lane at Pahnones Playazo ! A. triquetrum L. Bushy places ; very common; 2-4. I.! III.! [A. vineale L. and A. itrsinmn L. reported to Kelaart from the Eock have never been confirmed.] ■•'-Nothoscordum fragrans Kunth. Sandy places ; occasional ; 4-6. Flowers creamy white, with a red band ; no alliaceous odour. I. Bruce's Farm ! III. i. Campamento Village ! ii. Palmones Playazo ! COLCHICACE^. Colchicum Bivonce Guss. (C. autunmale L. var. gibraltaricuiii Kel.). Eough slopes; locally frequent or common? 9-10. 1. Middle and upper slopes ! III. i. Behind Campo Cemetery ! JUNCACE^. Juncus acutus L, Sandy or muddy ground, chiefly near sea ; common ; 3-5. Very pungent and rigid, heads terminal, dense, capsules very large. I., Ayala. By Sentry Fence ! II. ! III. ! Malaga Gardens ! J . maritimus Lamk. Similar places; locally abundant; 5-7. Tall and rigid, heads terminal, much laxer and longer, capsule smaller. III. i. Linea, K. ii. Palmones Playazo ! About Alge- ciras ! iii. Salt Pans ! Aguacorte ! /. stthulatus Forsk. (/. multiflorus Desf.). Muddy places near sea ; locally common ; 5-7. Leaves not septate, inflorescence lax, flowers not in clusters. III. ii. Behind Sandy Bay ! Near Eeina Cristina Hotel ! iii. By Eailway ! Tidal Elvers ! Salt Pans I Var. or sp. afl". My 2200, with short capitate inflorescence, from marsh near Algeciras, may be a new species allied to this ! J. Tenageia L. fil. Marshy and sandy places ; locally common ; 4-5. Annual, slender, inflorescence very lax, capsules small, sub- globose, dark brown. III. i. Hills near San Eoque, Ball ! ii. Marshes in Waterfall Valley ! iii. Palmones Sands ! J. bufonius L. Damp roadsides, streams, &c., very common ; 4-5; III.! Var. fasciculatus Koch, with two or more flowers in a cluster, is commoner than type. II. Eare, K. III. ! t/. foliosus Desf. Similar places ; rare '? 4-5. Leaves more numerous, broader, 1 line or more, sepals with two black lines on back. III. ii. Marsh south of Waterfall Valley, and in a valley a mile north of it ! J. glaucus Ehrh. By streams ; occasional ; 5-7. Wiry, stems leafless, inflorescence lateral, basal sheaths very shining, dark 108 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD brown. III. i. By Lajo ! Almoraima Soto ! ii. Below EI Cobre ! Near Las Corzas ! Carnero Hills ! iii. Salt Pans ! J. effusus L. Similar places ; rare ; 5-8. Habit of last, but soft, stems finely striate, flowers greenish, usually in a diffuse panicle, capsule not mucronate below style. III. ii. At and beyond Waterfall ! Var. compactus Hoppe has dense compact inflorescence. III. ii. Below El Cobre ! /. conglomeratus L. Similar places ; rare ; 5-8. Near last, but stem finely striate and rugulose, inflorescence dark, compact, and capsule mucronate below style. III. ii. Below El Cobre ! /. ohtusiflorus Ehrh. Marshes and by streams ; frequent, locally abundant ; 5-6. Stem leafy, leaves septate, inflorescence with very divaricate branches, perianth segments whitish, all obtuse. III. i. By the Cagancha ! Near First Venta ! By Lajo ! Cork Wood Sotos ! ii. Palmones Playazo ! iii. Guadacorte Marshes ! 1/. sylvaticus Keichb. fil. Similar places ; rare '? 6. Leaves septate, panicle branches ascending, perianth segments all very acute, pale. III. ii. Near Waterfall ! fVar. confertus Lge. has heads more compact. III. ii. Valley above Waterfall ! J. lamin-ocarims Ehrh. Similar places ; occasional or frequent ; 5-7. Like last, but branches ascending, perianth segments very dark, outer acute, inner obtuse. I beheve more general than my records show. III. i. Lajo below First Pine Wood ! ii. ByLobo behind Sandy Bay ! J. striatus Schousb. Similar places and in fields ; rather common ; 5-7. Stems subsolitary, short, leaves septate, much compressed and striate, inflorescence blackish. Like last, but all perianth segments acuminate, with spreading tips. II. ! III. ! /. Fontanesii J. Gay. Marshy places; occasional or frequent ? 5-7. Habit of last, but usually in larger patches, less leafy, perianth green or reddish, with long acuminate tips. III. i. By Lajo ! Marsh near Second Venta ! Seen, I think, in several other places, but no records kept. f /. supinus Moench. Similar places ; locally frequent ; 6-8. Dwarf, caespitose, leaves filiform, septate, heads small. Very vari- able in habit. III. ii. Waterfall Valley ! iVar. uliginosiis Both, is a lax procumbent rooting form. III. ii. With type ! \J. pygnicsus Eich. Sandy ground ; very rare ; 4-5. Annual, very small, stem-leaves 1-3, glomerules several, perianth segments green, adpressed. II. ! f /. capitakis Weig. Similar places ; occasional or frequent ; 4-5. Dwarf annual, stems leafless, glomerules sohtary, perianth segments dark, tips spreading, very acute. III. i. Bond's Farm ! Near Punta Mala ! Almoraima, P. L. ii. Near Algeciras, Nilss. S. de Palma, Bev. jLuzula Forsteri DC. Woods ; locally frequent ; 2-3. III. i. First Pine Wood I Cork Woods ! ii. M. de la Torre ! Mountains ! a flora of oibraltah and the neighbourhood 109 Cyperace^. Cypems capitatus Vand. {C. schosnoides Griseb., Schcenus mucro- natus 1j.). Sand-dunes; locally common ; 4-6. I. Catalan Bay ! North Front ! II. ! III. i. Linea ! ii. Sandy Bay ! Palmones Playazo ! iii. Palmones Sands ! G. rotunclus L. (C olivaris Targ.-Tozz.). Eoadsides and waste places; rather frequent ; 4-12. I. Below Lunatic Asylum ! North Front ! In the town, K. II. K., D. III. ii. Algeciras Station and south of the town ! iii. Sands by Palmones Kiver, Boiss., Wk. Salt Pans ! Guadacorte ! ■'• G. escidenkis Ii. Similar places ; rare? 7-9. Taller, spikelets paler, shorter, distichously set. III. ii. Sea sand by Palmones River and Algeciras, Rev. G. longus h. Wet places ; type rare; 3-8. III. i. Almoraima Soto ! A form with few, small, pale spikelets. Var. hadius C. B. Clarke. A tall stout plant, very common. II.! III.! ■\Pycreus Mmidtii Nees {Cyperus Eragrostis Wk. non Vahl., G. turfosus Salzm., G. pallescens Deb. non Desf.). Marshes ; rare ; 5-7. Short, leafy, leaves shorter than stem, the involucral shorter than rays, rays 4-5 long and 2-3 subsessile, spikelets rather dark, lanceolate. III. i. Alcadeza Crags ! Soto Gordo ! ii. Salt marsh near Algeciras, Eev.\, erroneously named Gyperus pallescens Desf. iii. Near Palmones, Bev., probably this. P. flavescens L. {Gyperus Gussonei Gasp.). Marshes ; rare? 5-7. Annual, tufted, short, slender, spikelets pale, subcapitate. III. ii. Algeciras, Bcv. P. globosus Reichb. Damp sandy places ; very rare ; 6-8. Heads dense, globose, whitish. III. i. A single clump near ford between San Roque and Pinar de los Bigotes ! Eleocharis palustris R. Br. Pools and wet places ; common ; 3-6. II.! III.! f^. mnlticaulis Sm. Damp heathy spots; occasional or locally frequent ; 3-6. III. i. Queen of Spain's Chair ! ii. Waterfall Valley ! Schcenus nigricans L. Similar places ; locally common ; 2-5. III. i. Foot of Chair both sides ! Arroyo Viejo ! Alcadeza Plains ! ii. By Miel ! Beyond Carnero Point ! Scirpus cernuus Vahl. [S. Savii Seb. & Maur.). Streams and wet places ; rather frequent ; 3-7. III. i. Cork Wood Sotos ! ii. Palmones Pinar and Playazo ! Mountains ! iii. Salt Pans ! Guadacorte Marshes ! Var. Valilii Lge. has larger solitary heads, with shorter bracts. III. ii. Near Algeciras, Fritze. S. lacustris L. Marshes ; locally plentiful ; 4-7. III. i. Be- tween Almoraima and Long Stables ! iii. Guadacorte Marshes ! S. sp. Pools ; very local ; 5-7. Stems very tall, pale, sub- cylindrical, inflorescence diffuse, branches 7-8, very unequal, each with 3-8 unequally pedicelled heads, umbellately set, 2|-3 lines long. Near S. Uttorale Schrad., but stem not triquetrous. Pro- 110 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD bably a new species, but my speciinens are too young to describe fully. III. i. Campo Common ! S. Holoschanus L. Sandy places ; very common ; 4-6. II. ! III. ! Var. axistralis Koch has smaller fewer heads, and grows with the type. III. ! S. maritimus L. Pools and ditches, and by rivers ; very common ; 3-6. II. ! III. ! Var. compactus Eeichb. fil. is a form with sessile heads, hardly worth distinguishing. It grows frequently wnth type. II. ! III. ! \Fuirena pubescens Kunth. By streams ; rather rare ; 5-6. III. i. East slopes Chair! ii. Waterfall Valley ! Slopes towards El Saladillo ! S. de Palma in Los Barrios district. Rev. \Rynchospora glauca Vahl var. pauciseta Turrill in Journ. Bot. 1914, p. 14. Wet places ; very local ; 6. Caespitose, with lax elongate inflorescence, on very long flaccid stems. An acquisi- tion to the Flora of Europe. The variety also occurs in Algeria (sp. herb. Kew. labelled P. laxa E. Br. !). The type is widely spread. III. ii. Above Waterfall ! Carex divisa Huds. Sandy grassy places ; frequent ; 1-5. I. North Front! II. ! III. ! Eare north of San Eoque. C. vulpina L. Bushy places and by water ; frequent or com- mon ; 3-6. II.! III.! G. divulsa Good. Banks and bushy places ; frequent ; 3-5. I. Below St. Bernards. III. ! Some forms look like C. muricata L., which is recorded from the neighbourhood by Pourret {ex Colm.), but I think ours is all C. divulsa. \G. panicidata'L. Swamps; locally common ; 1-4. III. i. In most of the Cork Wood Sotos ! Below S. Lorca ! C. distachya Desf. {C. Linkii Schk.). Dry bushy places; frequent? 2-4. Spikelets mostly androgynous, some female on long slender stems from base, utricles glabrous, stems 12-18 in., leaves J-^ line wide. I. Slopes below Middle Hill ! Frequent on Windmill Hill ! ? Perhaps C. Halleriana. III. i. Queen of Spain's Chair ! ii. Palmones Pinar ! The species has been confounded with C. ambigua, C. depressa, and C. Halleriana. One of the four occurs in Cork Woods, Alcadeza and Herring Bone Crags, M. de la Torre, and in the Waterfall Valley ! \C. ambigua Link. Similar places; rare? 2-4. Like last but shorter and less erect, leaves |-1 line broad. III. i. Queen of Spain's Chair ! \C. acuta Fr. By streams; locally common; 3-4. III. ii. By Miel above Algeciras ! Also, I think, in some neighbouring streams. C. glauca Scop. Damp, usually clayey spots on hills ; the type rare ; 3-5. III. ii. Near upper aqueduct, Algeciras ! Var. sermlata Ball, with less exserted peduncles, and narrow fruit tapering at both ends, is common. III. i. and ii. ! C.pendula Huds. (C maxima Scop.). Wet places in woods ; locally frequent; 4-5. III. i. Cork Wood Sotos! ii. Mountains! Hills near Carnero Point ! A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 111 C. hispida Willd. Damp places ; frequent ; 4-5. III. ! Var. anacantha G. & G. has very acute but hardly acuminate glumes. III. iii. Guadacorte Marshes ! (1399). iC. depressa Link (C. hasilaris Jord.). Dry bushy places ; rather rare ; 2-4. Resembles C. ambigua, but spikelets all unisexual, and utricles slightly pubescent. III. i. Queen of Spain's Chair ! ii. Hills above Algeciras, Ball. There is a specimen from Algeciras at Kew without collector's name! C. Halleriana Ass. Similar places ; rather rare ; 2-4. Like last, but taller and more slender, with pubescent utricles. I. Mediterranean Road ! III. i. Queen of Spain's Chair ! G. distans L. Marshy places; very common; 4-6. III.! C. B. Clarke unites this with G. Hornschuchiana Hoppe, which I think also occurs. Var. bcetica Auers. has green, not fuscous, glumes. III. i. About San Roque and Gibraltar, Wk. ii. S. de Palma, Bev. iC. binervis Sm. Similar places ; occasional; 3-6. Like last, but utricles with two strong green lateral ribs. III. i. Almoraima Soto ! ii. S. de Luna, Nilss. iii. Guadacorte ! ■\G. punctata G&nd. Marshes; rare; 2-4. III. i. Soto Gordo! iG. extensa Good. Marshes near sea; locally common; 4-6. III. ii. Palmones Playazo ! iii. Palmones Marshes, a form with very remote lower spikelets ! fC. Icevigata Sm. Damp woods and by streams; locally fre- quent; 3-5. III. i. Almoraima and other Sotos! ii. Mountains to top of El Frayle Ridge. Near Pelayo I Gramine^. \Leersia hexandra Sw. Swamps ; rare ; 6-7. III. i. Spring above Pindalista ! ii. ? Near Algeciras, Bev. Probably same station as next. iii. Guadacorte Marshes! Phalaris brachystachys Link. Cornfields ; very common in Spain ; 4-6. Annual, spikes about twice as long as broad, both scales below florets very short. I. Above Alameda ! III. ! Probably the species recorded by Masson as P. canariensis L. P. minor Retz. Dry waste places ; probably common ; 4-6. Like last, but spikes longer, one scale below florets one-third length of pale. I. Reclamation Road ! Near Willis's ! III. ! P. paradoxa L. Cornfields ; rather frequent ; 4-6. Spikes usually half enclosed in upper sheaths, some or all spikelets aborted and indurated. III. ! P. carulescens Desf. (P. bulbosa Cav. non L.). Damp clayey spots ; common ; 4-6. Tall, perennial, with bulbous root. III. I Var. major Wk. (sub P. bulbosa) is the taller form, often 4 ft. or more ! Anthoxanthum ovatum Lag. Dry fields ; abundant and very variable ; 4-5. III. ! \Plileum pratense L. Dry fields; rare; 4-6. I. K. III. i. S. Carbonera, D. fVar. nodosum Gaud, is decumbent, with swollen lower nodes 112 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD and short slender spikes. III. i. S. Carbonera, D. ii. North of El Cobre ! Crypsis aculeata Ait. Dry beds of pools ; locally frequent ? 6-8. Heads hemispherical or shortly oblong. III. iii. Excava- tions by railway ! Heleochloa {Crypsis) schomoides Host. Similar places ; rare ; 6-8. Heads oblong or cylindrical. III. iii. Eoadside excavation, Guadacorte ! Tragus racemosus Scop. Sandy and cultivated fields ; rare ; 4-6. III. i. Near Gibraltar, Cabrera, Clem. Digitaria sancjuinalis Scop. Cultivated soil; rare; 6. III. iii. Garden at Salt Pans ! ■'•Paspalum Digitaria Poir. Sandy shores ; rare ; 6. II. West shore ! Setaria verticillata P. de B. Cultivated soil ; rare ; 6-9. Awns with deflexed scabridity, setae two. I. Alameda Gardens ! and elsewhere, D. III. ii. Eoadside near Algeciras Casino ! S. viridis P. de B. Similar places ; rare ; 6-7. Awns with erect scabridity, setae several. III. i. Cultivated ground at San Roque, D. Panicum repens L. Sandy ground; rather common; 5-11. I. North Front ! II. ! III. ! A dwarf form, almost covered with sand, occurs in patches in sand dunes north of Linea ! Echinochloa Crus-Galli P. de B. Cultivated ground ; frequent ; 5-9. III. i. and ii. ! iii. ? Andropogon distachyum L. Bushy ravines ; rare ; 4-9. III. i. Queen of Spain's Chair! Majaram bout Crags! Alcadeza Ravine 1 ii. S. de Saladillo, P. L. Cymhopogon [Andropogon) Mrtus Stapf. Dry fields; abundant in Spain ; 1-12, but chiefly 4-6. I. ! III. ! Var. longiaristatus Wk. with awns 1 in. long or more, and sheaths glabrous below panicle, is as common as type ! tVar. podotrichus Hack, with spikelets longly villous above, the hairs on tubercles. III. ii. Algeciras, Rev. \Clirysopogon Gryllus Trin. Dry places; rare; 6-7. Super- ficially like last, but spikelets ternate on many long slender branches from each node, in a denser panicle. I. K. III. i. Near Gibraltar, Von Martins. Sorghum halepense Pers. Cultivated ground ; occasional ; 6-9. A stout broad-leaved species, with lax pyramidal panicle of sub- spiciform branches, leaflets with a broad white midrib. III. i. In maize fields by the Lajo ! Railway near Second Venta ! ii. About Algeciras aqueduct! iii. Gardens at Salt Pans ! hnperata cylindrica Pers. Grassy places ; very local ; 5-7. Panicle narrow, spiciform, becoming very white-silky. III. iii. Guadacorte Marsh ! Cynodon Dactylon Pers. Sandy grassy places ; very common in Spain, less so on the Rock; 1-12, but chiefly 5-9. I. Scattered about lower levels ! II. West shore ! III. ! Spartina stricta Roth. Muddy salt marshes ; local ; 6-9. III. Near San Roque, Massonl Probably same station as next. ii. and iii. Guadarranque Marshes, both sides of river 1 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 113 Arundo Donax L. Sandy places, partial to clamp ; common ; but almost always planted for fences; 9-11. Glumes subequal, lower pale with long silky hairs below. [I. K.] III. In all sub- districts, but looking native only by R. Lobo ! Phragmites communis Trin. By shore and by rivers ; rather frequent ; 7-9. Glumes very unequal, lower pale glabrous. I. Sandy Bay I III. i. Along the shore ! By the Lajo in many places, ii. By the shore ! iii. Frequent by Aguacorte Eiver and elsewhere ! Often with barren shoots several yards long. P. gigantea J. Gay, 12-16 ft. high, is commoner in the province, and may have been confounded with it or with Arundo Donax. I have seen none of them in flower. Ammophila arenaria Link. var. arundinacea Husnot {Psamina australis Mab.). Sand dunes; rather frequent; 5-6. More slender than type, with longer hairs at base of pales. II. East shore ! III. i. Linea Sands ! Near Campamento Common ! Guadarranque ! ii. Palmones Playazo ! Sandy Bay! iii. Pal- mones Sands ! Agrostis Juressi Link. Grassy places ; rare ; 6. Aspect of A. verticillata but annual, panicle contracted, subinterrupted, branches naked in lower part, pales unequal, the lower as long as glumes. III. ii. S. de Palma, Bev. By the Miel above and below Waterfall ! A. Benteri Boiss. Similar places; rare; 6-8. Perennial, panicle always very lax and slender. III. i. Campamento Com- mon, behind Fernando's ! A. alba Schva,d. Dry grassy places ; occasional; 5-6. Peren- nial, leaves flat, panicle lax or narrow. III. i. About Campamento and San Roque ! Var. gigantea Mey. is a large form with larger laxer panicle. III. i. By the Cagancha in several places ! Stream near Second Venta ! \N2iV. fuscescens Hack, has lanceolate acute glumes, becoming brown at maturity. III. ii. Algeciras, Bev. iVar. myriostachys Hack, resembles A. Beuteri, glumes linear- lanceolate, deep brown. III. ii. Algeciras, Bev. [Var. densiflora Pari. {A. scahriglumis B. & R.) is said to be frequent in the province.] A. verticillataNWl. By streams; locally rather frequent ; 5-8. Perennial, panicle dense, usually narrow, branches flowering to base, pales equal, shorter than glumes. III. i. Well in Linea ! By Almoraima Station, with rather lax panicle ! ii. Below Palmones New Bridge, a very large form ! Above Waterfall, and elsewhere in the mountains ! A. castellana B. & R. Dry hill slopes; the type rare; 5-6. Like A. alba, but lower pale with two minute lateral awns, and, in the type, a dorsal awn. III. i. Near San Roque, Ball ! Var. mutica Hack, has lower pale awnless, and is the usual form. III. i. Queen of Spain's Chair ! mixed with type. First Pine Wood! Campamento Common! ii. Waterfall Valley ! Journal of Botany, August, 1914. [Supplement] i 114 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD Var. tricuspidata Hack, is a form of the above. III. ii. Various places about Algeciras, Rev. A. setacea Curt. Dry hill slopes ; locally common ; 5-8. Leaves quite setaceous. III. i. Summit of Chair ! ii. Plentiful on slopes beyond the Waterfall ! A. iMllida DC. Dry places; abundant; 5-6. Annual, with very lax pale green panicle, and only one pale. III. ! A specimen from Porta k Rigo is labelled A. Cupaniana. A. interrupta L. Sandy places; rare; 6-7. Annual, with narrow interrupted panicle, and very long awns. I. ? Cle7n. Perhaps not on the Rock. Sporoholiis pungens Kunth. Sandy places near sea ; rare? 7-9. Stem with many distichously set stiff leaves, one fertile flower. I. ! ? II. ! ? III. i. Linea, D. ii. Algeciras, D. iii. Palmones, D. Flowers too late for my observations, but I have seen leaves apparently of this species at the Lighthouse, on the Neutral Ground, and at the mouth of Palmones River. There may be confusion with j^lurop7is when not in flower, but inflorescence very different. Gastridium lendigerum P. de B. Very common in cornfields, occasional in woods ; 5-7. I. K., Boiss. III. ! Polypogon monspeliensis Desf. Damp and waste places ; very common; 4-7. I.! II.! III.! P. 7?ian^M?ms Willd. Similar places ; frequent; 4-7. Usually smaller and reddish, glumes rather deeply bilobed. I. By Sentry Fence ! Queen's Road, at 700 ft. ! II. ! III. i. Campamento Common ! ii. Carnero Hills ! Estuary by Reina Cristina, Alge- ciras ! iii. Palmones Sands and Marshes ! Chceturusfasciculakis Link. Dry open places ; locally common ; 4-5. III. i. All round foot of Chair ! iii. Guadacorte Marshes ! Lagurus ovatus L. Sandy, waste and gravelly places ; very common ; 4-6. I. ! II. ! III. ! Stipa tortilis Desf. Dry open hills ; rather frequent ; 4-5. I. Mediterranean Road, &c. ! Europa Flats ! Buena Vista ! III. i. Cork Wood and Alcadeza Crags ! Both sides First Pine Wood! Campamento Common ! ii. Hills near Cortijo Trinidad! iii. Sands at Palmones, Bev. Macrocliloa tenacissima K-anih. Dry bushy hills and woods ; local; 4-6. I. Abundant by Charles V.'s Wall and above Levant ! occasional elsewhere ! III. i. Cork Wood Crags ! S. Carbonera! San Roque, Masson. M. areiiari a Kunth. Similar places; occasional; 4-5. III. i. East slopes Chair ! Cork Woods ! Alcadeza Crags ! ii. Water- fall Valley ! Piptatherum ccerulescens P. de B. Dry bushy slopes; very common on Rock, less so in Spain ; 4-6. Ligule long, branches of panicle few, spikelets 3|^-4 lines, awn of pales not longer than glumes. I. ! III. i. and ii. ! P. miliaceum Coss. (P. multiflorum P. de B.). Similar places ; common ; 5-6. Ligule short, branches of panicle many, spikelets 1^-2 lines, awn of pales about twice glumes. I. ! III. I A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 115 I Var. Thomasii (Kunth) has many of the lower panicle branches devoid of spikelets, but runs into type. I. Here and there mixed with type ! III. ii. The prevailing form in the Waterfall Valley ! Airopsis cjlobosaDesv. Sandy hills; rather rare? 4-5. Spike- lets small, very globose. III. i. Between Pinar del Eey and Jimena, Porta £ Bigo ! perhaps beyond our limits, iii. Between Algeciras and San Koque, Winkl. iAntinoria agrostidea Pari. Sandy turfy fields ; locally abun- dant; 3-4. A short caespitose perennial, spikelets small and short, one floret sessile, the other stipitate, glumes dark, longer than florets. III. ii. Pastures near El Cobre ! iii. Los Barrios Aviation Ground ! I believe also in i., but no definite record. Molineria minuta Pari. Similar places ; rare ? 3-4. Annual, panicle very lax, lower pale many-nerved, keeled, glumes shorter than florets. III. i. Gibraltar, Cabrera, Clem. Doubtless the neighbourhood is meant. Pinar del Eey, Porta d- Bigo I labelled Aira lendigera Lag. var. mutica Boiss. Var. hatica Wk., with hairs at base of lower pale aslong as the pale, is the commoner form in the province. III. ii. Near Algeciras, Hack. \Aira caryophijllea L. Sandy fields and open woods ; rather rare ; 4-5. III. i. San Eoque, Masson ! Cork Woods above San Eoque, Boiss. ! ii. S. de Palma, Porta & Bigo ! labelled Periballia hispanica. iA. multiculmis Dum. Similar places; rare; 4-5. Distin- guished from last by its clustered spikelets. III. i. Cork Wood Crags ! form with one-awned spikelets. There are two sheets at Kew from Boissier and Eeuter, one labelled A. capillaris Host., the other A. viuUicuhnis Dum., both "Gibraltar," which belong here. They probably come from San Eoque, not from the Eock. A. elegans Gaud. Dry sandy and stony places ; frequent, at least locally ; 4-6. Panicle large and very lax, only one flower awned. III. ii. M. de la Torre ! Plentiful in Waterfall Valley! [Var. biaristata G. & G., with both florets awned, is common in the province.] Gorynepliorus canescens P. de B. Sandy ground ; local ; 5-6. Perennial, in compact tufts, apical portion of awn gradually enlarged upwards, anthers 3-4 times as long as broad. III. i. Cork Woods near San Eoque, Boiss. iii. Palmones Sands ! (1080 and 2072). fC. fasciculatus B. & E. Sandy ground; occasional; 4-5. Annual, in lax tufts, awn as last, anthers quadrate. II. P>. III. i. Linea, D. Pine Woods near San Eoque, B. cC B. C. macrantherus B. & E. Similar places ; occasional; 4-5. Like last, but spikelets larger, in larger denser fascicles, apical portion of awn abruptly enlarged upwards, anthers linear. III. i. San Eoque, Boiss. Near Soto Gordo ! (2146). Almoraima, Porta & Bigo ! Deschampsia flexuosa Griseb. Grassy places in mountains ; / 2 116 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD rare ; 6-7. III. i. Queen of Spain's Chair ! ii. Mountains beyond Waterfall ! (2259). Dr. Stapf has only seen the latter specimen, which he considers inseparable from type. Var. oropJiila Hack, has a denser panicle with less spreading branches. III. ii. Summit of S. de Palma, Rev. Probably same station as mine. Avena sterilis L. Dry fields and waste places ; frequent; 4-6. Spikelets much larger than in next, only one floret disarticulating. I. Thinly scattered ! III. i. and ii. ! [Var. maxima P. L., a stout form, and var. scabriuscnla P. L., a more slender one, with lower part of awn scabrid, not villous, are said to be as common as type in province.] A. barbata Brot. Similar places ; common; 4-6. All florets disarticulating. I. ! III. i. and ii. ! A. sulcata J. G&y. Kough bushy places; rare; 4-6. Lower pale with 5-7 strong nerves, glabrous, with a tuft of hairs at base as long as pale. III. ii. S. de Luna, Winkl. A. albinervis Boiss. Similar places; locally frequent; 4-6. Lower pale much less strongly nerved, hairy in low^er third. III. i. Queen of Spain's Chair ! Majarambout Crags ! Almoraima, Porta d Iiigo\ labelled A. bromoides var. microstacluja Wk. ii. Slopes beyond Waterfall ! A. bromoides Gouan. Similar places ; rare ; 5-6. I. Medi- terranean Steps ! Willis's ! i Arrhenathermii elatius M. & K. Dry bushy places and woods; frequent ? 5-6. III. i. Near San Roque, Brouss. ii. Near Algeciras, liev. iii. At Palmones, Rev. Perez Lara has not seen this or the variety in the province, and suspects that the next species may have been mistaken for it. iVar. bulbosum Gaud, differs only in the base being swollen into one or more superposed bulbs, and is the only form I have seen. III. i. Queen of Spain's Chair, frequent ! ii. About the Water- fall! ? Perhaps ^. jjaZ/fins has been mistaken for it in the last station. A. erianthum B. & R. Similar places ; rare, but common else- where in the province ; 4-6. Differs from last in larger spikelets, awn produced from base instead of from middle of pale, and per- fect floret longly hirsute all over except at apex. III. ii. Below Waterfall ! t^. iMllens Link. {A. Thorei Duby). Similar places; local; 5-6, Spikelets smaller, both florets usually fertile, awn above middle of pale. III. ii. Mountains above Algeciras, B. c& Fc. ! Waterfall Valley ! S. de Palma, Winkl. Trisetum Dufo^lrei B. & R. Sandy fields and woods ; occa- sional ; 4-5. Ligule a ring of hairs, spikelets 2-flowered, awn 2-3 times as long as setse of pale. III. i. Sandy woods near San Roque, B. i(: R. ! Cork Woods ! ii. Palmones Pinar ! (959 and 1907). Probably Reverchon's " near Palmones River." iii. Lane from Guadarranque to Guadacorte ! Var. lasianthum P. L. has a very short lacerate ligule, not reduced to hairs, larger 3-flowered awnless spikelets, and more A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 117 villous very acute glumes. III. i.? Near Gibraltar, Boiss. ii. Near Algeciras, Hack. Palmones Pinar ! (937). Koeleria jMiiicea Domin {Trisetum negleclum Kunth). Sandy and grassy places ; very common ; 4-6. Sheaths villous, panicle 14-4 in. Looking like a lobed Polypogon. I. ! II. ! III. ! \K. ■pumila Domin {T. 2)umilum Kunth). Similar places ; much less common ; 5-6. Sheaths puberulous, panicle spiciform, |-1 in., awn shorter than in last. I. Above Willis's ! (1049). III. Pro- bably common, but confounded with last. K. jMeoides Pers. Grassy and gravelly places ; very common on Eock, rare in Spain ; 4-5. I. ! III. i. By Lajo below Almen- dral ! iii. Guadacorte ! Palmones, Eev. iVar. hrachystachya Domin is a dwarf form. I. Gibraltar, Von Martins. I think common on gravel paths, as on Queen's Koad near Michael's ! \K. Salzmanni B. & K. var. valdeinlosa Domin {K. phleoides var. Hack.). Sandy ground; very local ; 5. III. ii. Palmones Pinar ! (939). Probably Eeverchon's "Palmones" station. \Holcus grandiflorus B. & K. Bushy places in open woods ; locally common ; 5-6. III. ii. Slopes above and below Waterfall! H. lanatus L. Grassy places ; common ; 5-8. Not creeping, awn hooked at tip, scarcely exserted. II. ! Ill, ! Var. argenteus Hack, non Lge., less puberulent, often sub- glabrous, is as common and probably the prevailing form. III. i. San Roque, Boiss. ii. Waterfall Valley ! tVar. tiiberosus Ball. Rhizome subtuberous. III. i. San Roque, Ball. H. mollis L. Similar places ; rare ; 5-8. Stoloniferous, awn straightish or curved, longly exserted. III. i. Near San Roque, Brouss. ii. Marshy spot in valley beyond Waterfall ! Glyccria fluitans R. Br. Wet ditches; occasional; 4-6. III. i. Lajo Valley ! Almoraima Station ! ii. Roadside at Alge- ciras Bull ring! Gardens at Palmones Pinar! ii. Gardens at Salt Pans ! Seen elsewhere, but no precise stations noted. [Var. plicata Griseb., with shorter spikelets and florets, is fre- quent in the province.] iG.loliaceaGodr. Similar or drier places ; rare; 4-6. Resembles L. percnne. III. i. Near Gibraltar, Amo. -\Atropis iberica W.-Dod in Journ. Bot. 1914, p. 14. Tidal rivers; locally common ; 5-6. III. ii. and iii. Palmones River, both sides, about the bridges ! A. sp. My 1069 from the salt marshes near the stone bridge in Guadarranque Marshes is probably a new species allied to A. festuccBformis Richt. The material is not sufficient to diagnose. ■\Schismus marginatus P. de B. {Festnca calycina L.). Sandy places; rare; 4-6. I. K. III. i. Near Gibraltar, Von Martins. Poa annua L. Grassy places, roadsides, &c. ; very common ; 1-12. I.! II.! IIL! P. hulhosa L. Similar places ; occasional ; 3-5. Perennial, with a bulbous root and glaucous foliage. III. i. Queen of Spain's 118 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD Chair ! Campamento Common ! Pine Wood Plains ! ii. Hills south of Algeciras ! Mountains behind Saladillo ! P. trivialis L. Damp and clayey places ; common ; 4-6. II. ! III. ! Briza minor L. Grassy places ; very common in Spain ; 5-6. I. ? K. No confirmation. III. ! B. maxima L. Similar places ; very common in Spain, frequent on theKock; 5-6. I.! III.! Melica minuta L. (ilf. aspera Desf.). Eocky and stony places ; common on the Rock, less so in Spain; 4-6. The type is usually short, up to 12 in., with rather small subsimple panicle, and involute leaves, but it varies greatly. I.! III. i. Queen of Spain's Chair ! ii. Carnero Hills ! Waterfall Valley ! Var. arrecta (Kunze) has more erect panicle branches, and more unequal glumes, the outer longer than the florets. I. East slopes among Cliamcerops, Wk. ! Middle Hill ! (483). Var. latifolia Coss. {M. major Sibth. & Sm.) is much taller, with broad flat leaves and a large panicle. It is the form of bushy places. I. ! III. ii. Waterfall Valley ! S. de Palma, Eev. Var. pyramidalis (Bertol.) has a spreading pyramidal panicle. \M. ciliata L. Similar places ; frequent on the Rock, occasional in Spain ; 4-6. Panicle spiciform, lower pale longly silky. I. ! III. i. San Roque, Masson ! Var. major Ball (i¥. MagnoUi G. & G.) is a large form with lobate and interrupted panicle. I. Above Willis's! Probably elsewhere. III. i. West slopes Chair ! ii. Carnero Hills ! [M. nutans L. is reported near Gibraltar by Cabrera {ex Colm.).] Cutandia (Sclerochloa) maritima Benth. Sand dunes near the sea; rather local; 4-6. Spikelets and panicle branches often divaricate, pales very acute, keeled. I. Beyond Catalan Bay 1 II. East shore ! III. iii. Palmones Sands ! Scleropoa (Desmazeria) loliacea G. & G. Roadsides and waste places ; locally frequent on the Rock ; 4-6. Panicle compact, with erect subsessile spikelets, pales obtuse, not keeled. I. Governor's Cottage ! Catalan Bay ! Glacis ! III. ii. Algeciras, Winkl. S. rigida Griseb. Similar places ; common ; 5-6. Panicle smaller, rigid, with spreading pedicellate spikelets, pales as last. I. ! III. i. and ii. ! ^birojMs (Dactylis) littoralis Pari. Sandy shores; occasional? 5-8. Leaves many, short, stiff, distichous, spikelets 5-11-flowered in a lobulate spiciform panicle. I. A'. III. i. Between Gibraltar and San Roque, D. \Dactijlis glomerata L. Rough bushy places ; very common ; 5-7. A most variable species, of which the type has not hitherto been recorded. I. ! III. i, and ii. ! Var. australis Wk. (D. hisjpanica Roth.) is shorter, with a denser subsimple panicle, lower pale with deeper, rounded, not acute lobes, with shorter mucro. It is as common as type. I. ! At foot of Gibraltar, Sahm. ! III. ii. Waterfall! A FLORA OF GIbRALTAR AND THE NEIGHHOURHOOD 119 [Var. juncinella Boiss. is a smaller, more slender variety, with more puberulent, emarginate lower pale, and is, I believe, frequent.] Danthonla decumbens DC. Chiefly in watercourses on moun- tains ; locally frequent ; 4-6. III. i. Queen of Spain's Chair ! ii. Frequent in mountains ! Molinia caridea^ioenGh. Watercourses; local; 7-8. III. i. North-east slopes Chair ! ii. Valley beyond Waterfall ! Cynosurus echinatus L. Bushy slopes ; common ; 5-6. I. ! III.! C. clegans Desi. Similar places ; very local; 5-6. III. ii. At the Waterfall 1 Lamarchia aurea Moench. Eoadsides, walls, and waste places; rather frequent ; 3-5. I. Walls in the town ! Dockyard! Engi- neer Road ! III. i. Campamento village ! ii. Walls in Algeciras! iii. Palmones village ! Vulpia Myiirus Gmel. Grassy and rough places ; frequent ? 4-5. Leafy to top, panicle long and narrow, upper glume not awned. III. i. and ii. ! V. sciuroides Gmel. Similar places ; common ; 4-5. Stem longly naked at top, panicle short, dense. Not always easily distinguishable from last. III. i. and ii. ! Var. longearistata Wk. (F. BroteriB. & R.) has more numerous florets, and a longer awn to lower glumes. III. i. Almoraima, Boiss. San Roque, D. ii. S. de Palma, Bev. V. ciliata Link. Sandy places; rare? 4-5, Panicle long and narrow, silky-villous, partly enclosed in uppermost sheath, pedicels very short, lower glume not more than one-fifth as long as upper, pales longly ciliate. I. K. II. On the Isthmus, Wk. ! III. i. S. Carbonera, D. Almoraima, Wk. V. uniglumis Dum. Similar places; rare (but abundant in the province P. L.) ; 4-5. Both glumes awned, the lower one-tenth as long as upper or less. Habit of Bromits madritensis. III. i. Near Gibraltar, Von Martius. V. geniculata Link. Stony and rough places ; very common ; 4-5. Very variable, I have seen it 4ft. high by the Lajo. I.! Massonl III.! fVar. conferta Coss. & Germ, is a dv/arf condensed form. I. Europa Flats I V. Alopecuriis Link. Sandy ground; very common; 4-5. Spikelets on shorter pedicels, larger, and with more numerous florets than in last. I. Cave Guard, K. II. ! III. ! Near Gibraltar, Salzm ! Masson ! Var. lanata Boiss. with subsiniple raceme and very silky- villous pales seems as frequent as type. II. ! III. ! f Var. sylvatica Boiss. has elongate strict raceme, spikelets with 7-9 florets, and awn twice as long as pale. III. i. Cork Woods above San Roque, Boiss. I believe I hiave gathered this by the Lajo below Second Pine Wood. Festuca ampla Hack. Grassy hills ; rare ; 5-6. Ctespitose, slender, leaves flaccid, setaceous, convolute. III. i. Queen of 120 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD Spain's Chair! (2157). Found there also by Eeuter, teste Hackel. ■\F. montana M. Bieb. var. altissima Hack. {F. exaltata Wk. non PresL). Stony slopes; very local; 6-7. A very large and tall species. III. ii. At and above Waterfall ! Hackel and Eeverchon have also found it there. F. ccBvulescens Desf. Stony and bushy slopes ; locally com- mon ; 4-6. Stem-leaves flat, the radical often setaceous, densely covered at base of stem with remains of old sheaths. III. i. S. Carbonera, Hack. Majarambout Woods ! ii. Plentiful slopes beyond Waterfall ! F. arundinacea Schreb. var. mediterranea Hack. Damp clayey spots; frequent; 4-6. Awn fi'equently not terminal. I. Europa Glacis, on dry rocks, rare ! III. i. Marshy ground near Pinda- lista! (1195). iii. By Los Barrios Station! (2064). Seen in many other places, but doubtful between this variety and next. Var. Fenas Hack. (F. intcrrupta Desf.) is probably as frequent as last. III. i. Lajo Valley near upper ford ! Gibraltar, Fritze. Almost certainly in this subdistrict. Bromus tectorum L. ? Eocky slopes ; rare or error ; 4-6. Panicle drooping, rather dense, spikelets smooth, awn about as long as pales. I.? D. Not confirmed. B. sterilis L. Eough grassy places ; occasional '? 4-6. Very like B. 7naximus var. Gussonei, but decidedly smaller, with shorter awns. I. Eeclamation Eoad ! Charles V.'s Wall ! and probably elsewhere. III. ii. El Cobre ! Doubtless much more frequent than these records show. B. maximus Desf. Sandy ground ; common ; 3-5. The type has pale green very shortly pedicelled spikelets, with very long awns. Either this or the next species is common, but their differences are obscure, and the two have been much confused. I. ! Specimens from Catalan Bay (884 and 1157) have fascicled bulbs at the roots, but Dr. Stapf thinks they are only a state of the type. III. ! Var. Gussonei Pari, is much more diffuse and lax, resembling B. sterilis. It appears to be the commoner form, at least in Spain. III. ! fVar. glaber Wk. is a dwarf, few-flowered variety, with glabrous and shining spikelets. III. i. Near Gibraltar, Wk. B.madritensisDesi. Similar places; frequent? 3-5. Smaller than last with shorter awns, spikelets usually purple. I. Brouss I Willis's Eoad ! III. No actual records, but it certainly occurs, and is abundant in the province. Var. ciliatus Guss. has a dense subthyrsoid panicle, with softly pubescent, not scabrid glumes. I. Catalan Bay ! B. ruhens L. Similar places; occasional; 3-5. Panicle denser, rigid, reddish, spikelets subsessile. I. South and west slopes, K., D. III. i. San Eoque, Boiss. i Serrafalcus commutatus Godr. Grassy places, partial to shade ; occasional or frequent ; 4-6. Panicle large and lax, often drooping, pedicels long. III. i. Almoraima Soto ! ii. Hills south of Alge- A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 121 ciras! iii. Guadacorte Marshes! and probably many other places. S. Iwrdeaceiis G. & G. (S. mollis Pari.). Grassy places ; very common in Spain, occasional on the Kock ; 4-6. I. Above Alameda! Buena Vista! By North Front Cemetery ! II.! III.! Var. leiostachys M. k K. is a glabrous form. III. iii. Pal- mones Marshes ! Var. contractus Lge. ? has a compact panicle with short pedicels. III. i. Pedrera ! *S. Cavanillesii Wk. [Bromus scoparius L.), Sandy ground ; rare; 4-6. Panicle very dense, awns long, divaricate. III. ii. Algeciras, Winkl. S. squarrosus Bab. Sandy and cultivated fields ; rare ; 5-6. Panicle very lax, spikelets pendulous, lanceolate, with very divaricate awns. III. i. Near San Roque, Brouss. S. viacrostaclujs Pari. Similar places ; frequent ; 5-6. Like last, but panicle and spikelets erect, the latter larger and broader. III. Var. hrcvispicatus Boiss. has smaller more simple panicle. III. i. By watercourses, Campamento Common ! Hordeum murmuvi L. Sandy and waste places ; very common ; 3-5. I.! II.! III.! Var. inajor Boiss. with inner glumes of sterile spikelets ciliate on both sides, appears common. I. Rosia ! III. ! H. maritimum With. Dry sandy and clayey spots ; very common, often far inland ; 5-6. I. Grassy rocks, D. II. ! III. ! H. hulhosum L. Chiefly by dry watercourses; very common; 5-6. III. ! ^gilops ovata L. Dry grassy and sandy places ; very com- mon, except on Rock ; 4-6. I. Windmill Hill ! Europa Flats ! Governor's Cottage ! III. ! ^. triaristata Willd. Similar places ; much less common ; 5-6. Spikes larger, awns much longer, more erect, those of lower pale very unequal. III. i. Campamento Common ! By Lajo below First Pine Wood ! ii. About San Bernabe ! iii. About Palmones Bridge ! Guadacorte ! and elsewhere. jE. triuncialis L. Similar places ; occasional ; 5-6. Spikes linear, awns very erect. III. i. Campamento Common ! Lower slopes Chair ! Almoraima ! iii. Aguacorte ! Guadacorte ! No record for ii., but it doubtless occurs. Agropijron junceum P. de B. Sandy shores; locally frequent; 6-7. I. Governor's Cottage! North Front Butts! Catalan Bay ! Here forms are found with leaves flat, l|-3 lines broad, or involute, l-l^- lines when flattened. They look very different. II. ! III. i. Sands near La Tunara ! iii. Palmones Sands ! A. elongatum P. de B. Muddy salt marshes ; rare ; 6-7. A stout, stiff, erect species, with large spikelets. III. iii. Guadar- ranque Marshes ! f^. campestre G. & G. Sandy grassy places; frequent; 6-7. A very glaucous species, like a slender A. rcpcns. III. i. and ii. ! [A. repens P. de B. is abundant in the province.] 122 A FLORA OF GIBRALTAR ANfD THE NEIGHBOURHOOD Brachypodium sj/lvaticum R. & S. Woods and bushy places, chieliy in mountains ; frequent ; 5-8. III. i. and ii. ! B. ramosum R. & S. var. phcenicioides Koch [B. innnatum var. australe G. & G.). Bushy and stony places; frequent or common ; 5-7. Forms dense mats of barren shoots, sometimes a foot thick and three or four broad. I. Above Alameda! Plentiful on southern slopes ! III. ! B. mucronatum Wk. Rough stony slopes ; locally common ; 5-7. In small tufts, with few or no barren shoots. III. ii. Slopes beyond Waterfall ! B. distachyum P. de B. Grassy places ; frequent ; 4-6. Annual, not perennial, as the last two. I. ! II. ! III. i. and ii. ! Var. immilum Wk. is a reduced form with 1-2 spikelets, each 5-10-tiowered. I. Catalan Bay ! Var. muUiflorum Wk. has 4-5 spikelets, each 12-24-tiowered. III. i. Campamento Common ! Lolium pcrenne L. Sandy fields ; very common ; 4-6. I. ! II. ! III. ! Var. tenue Coss. & Germ, is a slender form, with 3-4-flowered spikelets. II. K. III. i. Sand Desert, D., Salzm. L. multiflo7'um hamk. {L. italic um k. Br.). Grassy and sandy places ; occasional ? 4-6. Spikelets awned, usually annual. III. i. By upper ford on Lajo ! I think in many other places, but have no definite records. L. rigUhim Gaud. (L. strictum Presl). Sandy and grassy places ; rather frequent ; 5-6. Spikes very rigid. I. Catalan Bay ! A tall erect form like L. pcrenne. III. i. Almoraima ! Pindalista ! Alcadeza Crags ! iii. Guadacorte ! All decumbent forms, with curved subcylindrical spikes. L. teniulentum L. Sandy places and cornfields ; locally fre- quent ; 4-6. III. i. Campamento Common ! Bond's Farm ! First Venta ! Common north of San Roque ! ii. Occasional ! Var. macrochcBtum A. Br. is a long-awned form. III. i. By San Roque Bull Ring ! &c. Gaudinia fracjiUs P. de B. Grassy places ; very common ; 5-6. A form with glabrous spikelets occurs frequently. III. ! Nardiinis tenellus Reichb. Sandy and cultivated fields; rare; 5-6. A slender annual, with unilateral spikes. III. i. San Roque, Brouss. Lcpturus incur vatusTxm. Sandy ground ; rare? 5-6. Glumes two. III. i. Sea sand near Gibraltar, Salzm ! L. jiliformis Trin. Similar places ; frequent or common ; 5-6. Very near last, but spikes straighter, spikelets not longer than internodes, and glumes not longer than spikelets. II.! III.! L. cylindricus Trin. By paths and in cornfields; rare? 5-6. Glumes solitary. III. i. Gibraltar, Salzm ! Probably in the neighbourhood, ii. Algeciras, Bev. Equisetace^. Equiscttim Tcbnateia Ehrh. Chiefly damp bushy places in A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD 123 woods, but not confined to them ; frequent ; 2-3. III. i. and ii. ! E. ramosum Schl. Dry sandy places and cornfields ; very common ; 3-5. I. Lew. III. ! This is doubtless the species referred to by Kelaart (Fl. p. 187) as E. hyemale. [E. hyemale L. Sandy banks ; very local, or error ; 3-5. III. i. Abundantly by a tributary of the Guadarranque, K. Probably the Lajo is meant, where the last species is abundant.] IsOETEiK. ilsoetes bcetica Wk. In pools ; rare ; 6. III. i. Bare in pools at Almoraima, Wk. Lycopodie^. Selaginella denticulata Sprg. Kough slopes and rocky places among bushes ; very common ; 2-4. I. ! III. i. and ii. ! FlLICES. Gymnogramma Icptophylla Desv. Bushy and stony places ; very common ; 3-5. I. ! III. i. and ii. ! Ceterach officinarum Willd. On rocks, partial to limestone ; 11-5. I. ! III. i. Alcadeza Crags ! Long Stables Eavine ! ■\ Notochlcena Marantce R. Br. Similar places ; rare ; 4-5. Re- sembling last, but fronds much larger and bipinnate, glabrous above. III. ii. Algeciras Mountains, Clem. ■\N. veiled R. Br. [N. lanuginosa Kaulf.). Similar places; rare; 11-3. Smaller than last, woolly both sides. I. Lcm.\ Boiss., Nee. Poly podium vulgare L. Rough rocky places, often on tree trunks ; rather locally common ; 8-3. I. St. George's Hall ! Mediterranean Steps! III. i. Queen of Spain's Chair! Cork Woods ! Alcadeza ! ii. Common in mountains ! Var. serratum Willd. has lobes of frond serrate. I. Mediterra- nean Steps ! III. i. San Roque, Wk. ii. Algeciras, Wk. Cheilanthes odora Svv. Under rocks and bushes ; rare ; 2-5. I. About Michael's I Below Spur Battery ! III. i. East Slopes Chair! ii. Waterfall! \Polystichu7n Thelypteris Roth. Streams and marshes; locally frequent; 7-10. III. i. S. Lorca ! Cork Wood Sotos ! P. Filix-mas Roth. Woods ; rare ; 6-9. III. ii. Cuartel de las Corzas, P. L. S. de Palma, Bev., Clem. Cystopteris fragilis Bernh. Shady slopes on mountains; rare; 6-9. III. I. San Roque, D. ii. S. de Luna, P. L. S. de Palma, Clem. Aspleniwn Filix-fosmina Roth. Damp places in woods ; locally common ; 6-8. III. i. Cork Wood Sotos ! ii. Mountains ! A. lanceolatum Huds. Dry, but shady rocks; locally common; 4-9. III. i. Summit of Chair ! ii. Mountains ! A. Trichomanes L. Similar places; common on the Rock, less so in Spain ; 4-9. I. ! III. i. Alcadeza Crags ! Long Stables ! S. Carbonera, D. No record for ii. where it certainly should occur. 124 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD A. Adiantum-nicjrimi L., var. Virgilii Heufl. Bushy and rocky places, chiefly in woods ; frequent ; 4-9. Only the variety is re- corded, which differs from type in narrower, more spreading, more deeply incised segments, and much longer linear sori, but I think type is common. I. K., D. I have no note of having seen it on the Rock, but think it is an accidental omission. III. i. Cork Woods ! ii. Plentiful in mountains ! Carnero Hills ! Scoloj^endriiimHemio^iitisG&v. Damp caves ; very local; 2-5. I. Michael's Cave, Clem., K., D. In a cave on east slopes ! Below Governor's Cottage, K. Blechnum spicant Roth. Damp rocky places in woods ; local ; 6-8. III. ii. In and about Waterfall Valley, and a valley north of it! Pteris aquilina L. Dry sandy places ; very common, except on the Rock ; 7-9. I. A starved plant or two growing through the concrete of a water chute just beyond the Mount ! III. ! Adiantum Capillus -Veneris L. Damp rocks and by water- courses; frequent in Spain, rather rare on the Rock; 6-7. I. Lower Union Gallery ! Cave above North Front ! Caves on eastern shore, herb. Balestrino ! Below Mediterranean tunnels ! The Rock form has much larger fronds than that from the country round. III. i. and ii. Scattered over these subdistricts ! Davallia canariensis Sw. On tree trunks and rocks ; locally very common; 3-9. [I. K., but obviously in error.] III. i. Cork Woods ! ii. Mountains ! Osviunda regalis L. By streams ; locally common ; 5-9. III. i. Soto Gordo! Almoraima Soto ! ii. Waterfall and other valleys ! OpMoglossuvi lusitanicum L. Sea sand ; rare ; 1-3. III. ii. x\lgeciras, Glem. APPENDIX Additional Species and Varieties. P. 5. Last line but three should read " Hesperis matronalis Lam. var. laciniata Boiss. {H. laciniata All.). Rocky places ; rare ; 4-5. I. Lcm. ! " P. 28. Medicago ohscura. Add "/. muricata Urb., a spiny- fruited form, is reported by Colmeiro from III. ii. Algeciras." P. 31. After T. laj^paceum insert " T. carteiense de Coincy in Journ. de Bot. xiii. p. 163 is recorded by its author as a new species from the Carteian Hills. I do not know it, but it seems near T. lapixtceum." P. 55. After Anthemis maritima insert " Ormenis nohilis J. Gay. Sandy heaths; rare; 5-9. Resembles next, but perennial. III. i. Near Gibraltar, Clem.'" P. 57. Before Phagnalon saxatile insert " Phagnalon sordidum DC. Rocky places ; rare ; 4-6. Known from next by adpressed phyllaries. I. '? Lag. III. ii. Algeciras, Nee." P. 63. After Hyoseris radiata and details insert " var. elongata Huet de Pav. Achenes with an inner pappus of 2-3 setae, the marginal winged. I. Fissures of Rock, Wk. III. i. Walls of San Roque, JVk." P. 66. After Crepis corymbosa var. hatica insert : — " Andryala ragusma L. Dry bushy hills ; rare; 5-7. III. i. Near San Roque, Boiss. ex Cohn. Var. minor Lge. III. i. With the type, Brouss. ex Colm." P. 77. Add " [Chcenorrhinum origanifolium L. is recorded by Willkomm in Bot. Zeit. 1845, p. 742, as found by him on the Rock, but he enters it from N. Spain only in Prodr. Fl. Hisp.] " P. 79. After line 17 add " Pedicularis lusitanica Hoffm. & Link. Damp spots in bushes and by watercourses ; occasional ; 3-5. III. i. Queen of Spain's Chair ! ii. Mountains ! iii. Palmones, Bev. P. 80. Before Orobanche minor insert " 0. variegata Wallr. III. i. Top of Chair ! Not previously recorded for Spain." * P. 80. Before 0. Picridis insert " 0. Hanseleri Reut. var. del lid ens/'- var. nov. Filamentis paulo supra basim insertis, corolUe lobis copiose glanduloso-pilosis, acriter et longius dentatis. Beck MS. III. ii. Waterfall (my 370) ! " P. 80. Before 0. minor insert " 0. Hederce Duby. III. ii. Near Almoraima ! " * The determinations of some of my Orobanches were received from Prof. Beck too late for insertion in their place. 12G A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD P. 84. Add " [Ddomis tuherosa, reported by Boissier (Voy. p. 70) from near the Galleries on the Eock, is probably meant for Nepeta tuherosa, which occurs there.] " Corrections. P. 1. Nigella damascena. Delete " ? " after I. and under II., after K. add '' Bev." and delete "probably same station." De- beaux's citation is quite misleading. P. 5, line 1. For "smaller" read " larger." P. 22. Buta hracteosa. Eemove " ? " after I. Erase remarks after K. and add " Boiss." P. 26. Erase II., and for " Neutral Ground " read " North Front." P. 34. Lotus angustissinms. Add " ? " Delete " Young examples — here," and substitute " Probably the next has been mistaken for it." Delete " Boiss." who only records it from " between San Eoque and Estepona," probably beyond our limits. Debeaux's citation is, as in many cases, misleading. He may have seen a specimen of Boissier's from Algeciras, but I have no confirmation. New Stations. With the exception of the first named, and of the Orobanches, only those previously unrecorded for their districts are given : — Delphinium Staphisagria. I. Hurst ! Fumaria macrosepala. III. ii. B. tt B. Cistus albidus. III. ii. Lag. Tuberaria echioides. III. iii. Bev. Drosop)hyllum lusitanicum. III. iii. Bev. Silene littorea. III. iii. Fritze, Winkl. Arenaria emarginata. III. iii. ! Gerastium Boissieri. III. ii. Von Martins. Spergula arvensis var. glutinosa. I. Lam. Lavatera trimestris. I. Boiss., K. Ononis puhescens. III. ! Lathyrus sativus. I. K. Myrtus com- munis. I. Tournef., Ayala. Lythrum Grcefferi. I. K. Bupleu- rum gibraltaricum. III. ii. Vahl. Helminthia comosa. I. K. ! Andryala arenaria. I. Boiss., Wk. Jasionemontana {tyTpe). III. i. Boiss. Jasminum fruticans. III. i. Schott. Calystegia Solda- nella. I. Ayala. Echimn italicum. III. i. Navarro, Pourr. Symphytum tuberosum. III. i. Von Martins. Linaria viscosa. III. iii. Bev. Or obanche feet ida i. pus ilia 'Beck. I. Above Willis's ! III. ii. Hills near San Bernabe! 0. sanguinea Presl. I. Above Willis's ! Stigmas yellow in my specimens ; Beck describes them as red. 0. Picridis F. Schultz. III. ii. Palmones Playazo! ''ad minorem transiens," Beck. Palmones Pinar! Var. Carotce, Beck. I. Levant ! 0. minor Sutt. I. Ince's Fai-m ! Above Devil's Gap ! Plantago Lceflingii. III. i. Von Martinis. Doubtful Stations. There is no satisfactory evidence that their collectors intended to specify the Eock, as distinct from the neighbourhood, for the following, which are therefore doubtful records for I.: — Malcolmia lacera. Linum gallicum. Astragalus Epiglottis. Ornithopus A FLORA OF GIHRALTAH AND THK NEIGHBOURHOOD 127 roseus var. isthmocarpus. Lathyms augulatus. Daucus crinitus. Galium divaricatum. Kent rophij Hum hceticiim. Campanula dichotoma. Lithospermumfruticosum. Linaria cirrhosa. L. tri- phylla. Lafuentea rotundifolia. Doubtful Names. The following names are doubtful. Where " I." is added, the doubt is as to the correct identification of the species for the Eock, though they may exist elsewhere: — Erodiuvi Jacquinianum and var. bipinnatum. E. Salzmanni. Ulex aphyllus. Ononis diffusa I. Governor's Cottage. 0. Natrix var, media I. Medi- cago turbinata var. aculeata I. M. coronata. Fceniculum officinale I. Artemisia pontica. Senecio leucanthemifolius. Cynara Gardunculus. Thrincia maroccana I. Andryala laxiflora. Erythrcga Gentaurium I. Solanum vUlosum I. Alterations of Status. Remove t from Velezia rigida and add to Opoponax Gkironium. Remove [ ] from district cited or from the whole species if no district is here indicated. Retama monosperma I. (Several old records exist.) Eidolfia segetiim I. (Doubtless a former weed of cultivation.) Echium italicum. Symphytum tuberosum. Plan- tago Lceflingii (except I.). Add [ ] to Senecio petrceus and to district I. for the follow- ing:— Brassica fruticulosa (erroneous record). Anthemis arvensis (casual). Senecio petrceus, Gardmis nigrescens and Helminthia echioides (wrong determinations). Anarrhinum bellidifolium. (Von Martius does not record this species as stated by Kelaart, but he does record A. tenellus, doubtless in error for GJicenar- rhinum villosum.) Add '■' to Solanum sodomceum. Amaranthus chlorostachys. Chenopodium ambrosioides. Deletions. Silene cerastioides (record for I., probably not the Rock). Genista Hcenseleri (record for I., Kelaart did not record it). Ononis campestris (record for I., Talbot's records are worthless). Asperula hirsuta, and Coleostephus Myconis (records for I., Kelaart did not record them). Andryala arenaria (citation of Boissier for II., he records it "on the Rock "). Nepeta reticulata (certainly an error). INDEX OF GENEKA. Italics are used for synonyms ; those referred to another genus are not mentioned in the text. PAGE Acacia 39 Acanthus 86 Ace7'as.. 101 Achillea 55 Achyranthes 89 Adenocarpus 25 Adiantum 124 Adonis 3 ^gilops 121 iEluropus 118 iEtheorrhiza 66 Agave 103 Agrimonia 40 Agropyrum 121 Agrostis 113 Aira 115 Au-opsis 115 Ajuga 85 Alchemilla 40 AlisQia 98 Allium 106 Alnus 97 Aloe 105 Alsine 10 Alternanthera 89 Althaea 19 Amaranthus... 88, 127 Ammi 45 Ammophila 113 Amygdalus 40 Anacyclus 55 Anagallis 86 Anagyris 23 Anarrhinuui ... 76, 127 Anchusa 72 Andropogon 112 Andryala 66, 127 Anemone 3 Anethuni 47 Anthemis 54, 127 Anthoxanthum ... Ill Anthylhs 32 Antiuoria 115 Antirrhinum 76 Apium 46 PAGE Arabia 6 Arbutus 68 Arenaria ... 16, 18, 126 Arisarum 99 Aristida 40 Aristolochia 93 Armeria 87 Arrheuatherum ... 116 Artemisia 56, 127 Arthrocnemum ... 89 Arum 99 Arundo 113 Asparagus 104 Asperiila 50, 127 Asphodelus 105 Asplenium 123 Aster 54 Asteriscus 54 Asterolinum 86 Asterothrix 86 Astragalus 34,126 Astrocarpus 12 Atractylis 59 Atriplex 89 Atropis 117 Aveua 116 Ballota 84 Barkbausia 66 Bartsia 79 Bellis 53 Beta 90 Betonica 84 Bifora 49 Biscutella 7 Biserriila 35 Blechnum 124 Boiijeauia 33 Borago 73 Bourgaea 60 BrachypoiHum ... 122 Brachytropis 11 Brassica 6, 127 Briza 118 Bromus 120, 121 PAGE Bryonia 42 £M«ias= Snccowia 6 Buphthalmuin = Aste- riscus 54 Bupleurum 46, 126 Cachrys = Hippoma- rathrum 49 Cakile 8 Calamintha 82 Calendula 58 Callitriche 95 Calluua 69 Calycotome 24 Calystegia 72, 126 Campanula ... 68, 127 Capnophyllum ... 49 Capparis 8 Capsella 8 Cardamine 6 Carduncellus 59 Carduus 60,127 Caiex 110 Carliua 59 Carregnoa 103 Catalpa 71 Catapodium = Sclero- poa 118 Caucalis 48 Celtis 96 Centaurea 61 Centranthus 52 Centunculus 86 Cepbalanthera ... 101 Cephalaria 52 Cerastium 17, 126 Ceratonia 39 Cercis 39 Cerintiie 72 Cestrum 75 Ceterach 123 Chaenarrhinum 77, 125 Chaetonychia 43 Chaetuius 114 Chamaerops 99 A FLORA OP GIBRALTAR AND THE NEIGHBOURHOOD 129 PAGli Cheilanthes 123 Cheuopodium 90, 127 Chlora 70 Chrysanthemuui SG Chrysopogon 112 Cicendia 70 Cicborium 62 Cineraria = Sene- cio 57 Cirsium 60 Cisms 8, 126 Cladanthus 55 Clematis 1 Cleonia 85 Colchicum 107 Coleosteplius... 56, 127 Colocasia 99 Couium 49 Conopodiuni 45 CoQVolvulus 71 Conyza 54 Corbularia 103 Coriandrum 49 Coriaria 22 Coridotliymus ... 82 Coris 86 Cornicina 32 Coronilla 85 Corrigiola 43 Corynephorus 115 Cotyledon = Um- bilicus 43 Crambe 7 Crataegus 40 Crepis 66 Crithmum 47 Crocus 102 Crozophora 93 Crucianella 50 Crupina 61 Crypsis 112 Cuscuta 72 Cutandia 118 Cydonia 41 Cymbopogon 112 Cynara 60, 127 Cj'nogl'issum 74 Cynodon 112 Cynosmus 119 Cyperus 109 Cystopteris 123 Cytinus 97 Cytisus 23 Dactylis 118 Danthonia 119 Daphne 92 PAGK Datura 75 Daucus 48, 127 Davallia 124 Delphinium ... 1, 126 Deschampsia 115 Desmazeria 118 Dianthus 12 Digitalis 78 Digitaria 112 Diotis 55 Diplotaxis 7 Dipsacus 52 Dolichos 39 Dorycnmm^^on- jeania 33 Dorycnopsis 33 Drosophyllum 11, 126 Ecbalion 42 Echinochloa 112 Echinops 62 Echium 73, 126 Elfeoselinum 48 Eleocharis 109 Emex 90 Endymion 105 Ephedra 97 Epilobium 42 Epipactis 101 Equisetum 122 Erica 68 Eriobotrya 41 Erodium 21, 127 Erucastrura 7 Eryngium 45 Erythrasa 70, 127 Erythrina 39 Eudyanthe 13 EutVagia 79 Eupatoriiim 53 Euphorbia 93 E uxoliis = Kmav&n- thus 88 Evax 57 Fedia 52 Ferula 49 Festuca 119 Ficaria 3 Ficus 96 Filago 56 Foeniculum ... 47, 127 Fragaria 40 Frankenia 11 Fraxinus 69 Fritillaria 104 Fuircna 110 Journal op Botany, August, 1914. [S PAGE Fumana 10 Fumaria 4, 126 Galactites 60 Galega 34 Galium 51, 127 Gastridium 114 Gaudinia 122 Genista 24, 127 Geranium 20 Geropogou 64 Gladiolus 101 Glaucium 4 Gleditschia 39 Glyceria 117 Gnaphalium 56 Gomphocarpus ... 70 Gymnogramma ... 123 Halimium 9 Halogeton 89 Hedera 49 Hedypnois 62 Hedysarum 36 Helianthemum 10, 9 Helichrysum 56 Heleochloa 112 Heliotropium 72 Helniinthia ... 64, 127 Helosciadium 46 Herniaria 43 Hesperis 5, 125 Hippocrepis 35 Hippomarathrum 49 Holcus 117 Hordeum 121 nijmenostemma = Prolongoa 55 Hyoscyamus 75 Hyoseris 63, 125 Hypericum 20 Hypochoeris 64 Iberis 7 Ilex 23 Illecebrum 43 Im])erata 112 Inula 54 Iris 101 Isnardia 42 Isoetes 123 Jasione 67, 126 Jasminum ... 69, 126 Juncus 107 Juniperus 97 UPPLEMENT.j k 130 A FLORA OF GIBRALTAR AND THE NEIGHBOURHOOD PAGE Kentrophyllum 59, 127 Kochia 89 Koeleria 117 Kohlrauschia 12 £'o?i/(7rt = Lobularia 8 Kundmannia 47 Lactuca Lafuentea ... 78, Lagurus Lamarckia Lamium Lathyrus 38, Laurentia Laurus Lavandula Lavatera 19, Leersia Lemna Leontodon Lepidium Lepigonuni Lepturus Leucoium Leuzea Limodorum Linaria 77, Linum 18, Lithospermum 73, Lobelia Lobularia Loeflingia Lolium Loniceva Lotus 83, Lupinus Luzula Lychnis (see Me- laudryuai and Eudyanthe) ... Lycium Lycopus Lysimachia Ly thrum 41, 65 127 114 119 83 127 67 92 81 126 111 98 64 8 17 122 102 61 101 126 126 127 67 8 43 122 50 126 25 108 13 75 81 86 126 Macrochloa 114 Magydaris 47 Malcolmia 5, 126 Malva 19 Mandragora 75 Marrvibium 84 Matricaria 55 Matthiola 5 Medicago 28, 125 Melandryum 13 Melia 20 Melica 118 PAGE Melilotus 29 Melissa 82 Mentha 81 Mercurialis 93 Mesembryanthe- mum 44 Microcala = Ci- cendia 70 Microlonchus 61 Micromeria 82 Mirabilis 88 Moehringia 16 Mcenchia 17 Moliueria 115 Molinia 119 Moniordica^^^cha- liou 42 Morns 96 Muscari 105 Myosotis 74 Myriophyllum ... 42 Myrtus 41, 126 Narcissus 103 Nardurus 122 Nasturtium 6 Neotinea 101 Neottia = Spiran- thes 101 Nepeta 83, 127 Nerium 70 Nicotiana 75 Nigella 1, 126 Nothoscordum ... 107 Notobasis 60 Notochlsena 123 Obione 89 Odontites 79 (Euanthe 47 CEnothera 42 Olea 69 Omphalodes 74 Onobrychis 36 Ononis 25, 127 Onopordon 59 Ophioglossum 123 Ophrys 100 Opoponax 49, 127 Opuntia 45 Orchis 99 Origanum 81 Orlaya 48 Ormenis 55, 125 Ornithogalum ... 104 Ornithopus ... 36. 126 Orobanche 79, 125, 126 PAGE Osmunda 124 Osyris 92 Otospermum ;= Matricaria 55 Oxalis 21 Pallenis = Aste- riscus 54 Pancratium 103 Panicum 112 Papaver 4 Parietaria 95 Paronychia 43 Paspalum 112 Passerina ^= Thy- melaea 92 Pedicularis 125 Peplis 41 Peridersea 54 Peris^^iMS = Orchis 100 Petroseliuum 46 Phaca 35 Phagnalon 56,125 Phalaris Ill Phelipaa 80 Phillyrea 69 Phleum Ill Phlomis 84, 126 Phragmites 113 Physanthyllis 32 Phytolacca 90 Picridium 65 Picris 64 Pimpinella 46 Pinardia 56 Pinguicula 86 Pinus 97 Piptatherum 114 Pistacia 23 Pistorinia 44 Pisum 39 Plantaso 87, 126, 127 Poa 117 Podospermum ... 64 Poly carpon 43 Polygala 11 Polygonum 91 Polypodium 123 Polypogon 114 Polystichum 123 Populus 97 Portulaca 43 Posidonia 98 Potamogeton 98 Poteutilla 40 Poterium 40 Prasium 85 A FLORA OF GIBRALTAR AND THE NEIGHROURHOOD 131 PAGE Prolongoa 55 Prunella 85 Prunus 41 Fsamma 113 Psoralea 34 Pteris 124 Pterocephahis ... 53 Pterospartuin 24 Piilicaria 54 Punica 41 Piipalia 89 Pycnocomon 53 Pycreus 109 Pyrus 41 Quercus 96 Radiola 18 Eauunculiis 1 Eaphanus 7 Rapistnioi 7 Reseda 11 Eetama 23, 127 Rhagadiolus 63 Rhamnus 22 Rhododendron ... 68 Ricinus 93 Ridolfia 46, 127 Rtemeria 4 Romulea 102 Rosa 40 Rosmarinus 82 EottboeUia= Scler- opoa 118 Roubieva 90 Rubia 50 Rubus 39 Rumex 90 Ruppia 98 Ruscus 103 Ruta 22, 126 Rynchospora 110 Sagina 16 Salicornia 89 Salix 97 Salsola 89 Salvia 82 Sambucus 50 Samolus 87 Sanicula 45 Saponaria 16 Sarothamnus 23 Satureia 82 Saxifraga 45 Scabiosa 53 Scandix 45 PAGK Schiuus 23 Schismus 117 Schoenus 109 Scilla 105 Scirpus 109 Sclerochloa 118 Scleropoa 118 Scolopendriuni ... 124 Scolymus 62 Scorpiurus 35 Scorzonera 64 Scrophularia 76 Sedum 44 Selaginella 123 Sempervivuni 44 Senebiera 8 Senecio 57,127 Serapias 99 Seriola 65 Serrafalciis 120 Serratula 61 Setaria 112 Slierardia 50 Sibthorpia 78 Sideritis 84 Silene 13, 127 Silybum 60 Simethis 105 Sinapis 6 Sisymbrium 5 Siuni = Kundmau- nia 47 Smilax 103 Smyruium 49 Solanum 74, 127 Sonchus 65 Sorghum 112 Sparganium 99 Spartina 112 Spartium 23, 24 Spergula 17, 126 Spergularia 17 Spiraea 40 Spiranthes 101 Sporobolus 114 Stachys 84 Statice 87 Stellaria 17 Stipa 114 Suaeda 89 Succowia 6 Symphytum ... 74, 127 Tamarix 42 Tamus 103 Tapeinanthes =^Ca.r- regnoa 103 PAGE Taraxacum 65 Teesdalea 8 Tetragonolobus ... 33 Teucrium 85 Thapsia 48 Theligonum 95 Thesium 92 Thrincia 63, 127 Thymehca 92 Thymus 81 Tolpis 62 Torilis 47 Trachelium 68 Tragopogon 64 Tragus 112 Tribulus 22 Trichonema=Ilomu.- lea 102 Trifolium 80, 125 Triglochin 98 Triodia=Da.ntho- nia 118 Trisetum 116 Trixago 79 Tuberaria 9, 126 Tulipa 104 Tunica 12 Typha 99 Ulex 24, 127 Ulmus 96 Umbihcus 43 Urginea 105 Uropetalum 105 Urospermum 04 Urtica 96 Vaillantia 51 Valeriana 51 Valerianella 52 Velezia 13, 127 Verbascum 76 Verbena 86 Veronica 78 Viburnum 50 Vicia 36 Vinca 70 Viola 11 Vitex 86 Vitis 20 Vulpia 119 Xanthium 67 Zannichellia 98 Zostera 98 Now York aolB 3 5185 00265 03: