ROYAL BOTANIC GARDENS, KEW. BULLETIN OF MISCELLANEOUS = INFORMATION. ONDON: PUBLISHED BY HIS MAJESTY’S STATIONERY OFFICE. To be purchased through any Bookseller or directly from HM. STATIONERY OFFICE at the following addresses: IupertaL Hovsr, Kinesway, Lonpon, W.C. 2, and 28, ABinepon STREET, Lonpon, S.W. 1; 37, Perer Street, MANCHESTER 1, St. ANDREW’s CRESCENT, CARDIFF; 23, FortH Street, EDINBURGH ; or from E. PONSONBY, Lrp., 116, Grarron STREET, Dustin. PRINTED BY JAS. TRUSCOTT anp SON, Lrp., Surrorx Lane, Lonpon, E.C, 4. 1918. Price 4s. 6d. THE SEPARATE NUMBERS OF THIS VOLUME WERE PUBLISHED ON THE FOLLOWING DATES :— a st es ..- March 13. Nos. 2 & 8 — in ews Aprils: sae el See =e oe ... dune 6. Neh “ ae mocemy &. 20s. 6 = a3 sid vee ... August 28. No. 7 ai a ut ... November 25. Nos. 8&9 .... + ~~» December 16. Bo. ld |: sim “Another species—R. racemosa, Meyer —is mentioned by him, but not described. : s long and delightful account of Rhizophora Mangle leaves no doubt as to the identity of my Form i. **Pedunculi communes axillis insident” solitarii ”” a writes, oo interdum triflori. et obsolete trigoni’’,t and of the on ‘seed ”? which he states is the longest’ part o of the fruit, longa”. Hi bait Ecicces americanae,” p. 14], pl. Ixxxix. + The italics throughont are mine.—A. L. 5 fruit is poor and contrasts unfavourably with his excellent repre- sentation of the inflorescence and flowers. It is evident that my Form i. is Rhizophora Mangle as defined by Jacquin. Meyer’s description of Rhizophora racemosa* is less satistac- tory. ‘‘Inflorescentia composita haec stirps insigni modo a But both my Forms ii. and ili. have compound inflorescences, and they are clearly distinct species. Of the cymes Meyer says “rami et ramuli crassi . . . quinquies vel sexies repetito- dichotom1 multiflori . . . in utraque dichotomia e margine calyciformi sexdentato enati’’, all of which agrees with m Form iii.; but he concludes with “‘ flores et fructus Lhizophorae Mangle, sed petala pilosa’. The flowers and fruit of my are elliptical or slightly ovate, instead of cuneate as in the case ot R. Mangle Mangle has a radicle which is frequently entirely red-brown. Moreover, the pericarp of my Form ii. is very stout and is often curved. e petals in fresh specimens of my Form 1. are Oliver, in describing Rhizophora racemosa from Tropical Africa, says ‘‘Flowers in pedunculate, divaricate, shortly-jointed, Binfais : more or less”?. In my Form iii. the joints of the cymes are distinctly shorter than in Form u., and sometimes may be * © Primitiae Florae Essequeboensis,” p. 185. 6 a Rhizophora wade var. racemosa but without discussing its peculiarities beyond the statement that its inflorescence is repeatedly a Cething Oliver remarks that R. racemosa ‘* occurs also in Tropical America, if this identification is correct’’.* It must be noted that Oliver describes . racemosa as “‘a ‘small tree’’, whereas in British Guiana, so far as I have seen, it is the largest of the three species I have noted. In British Guiana it is certainly a larger plant, in every’ way, than &. Mangle, which is described by Jacquin as attaining a height of ‘ quagintapedalem plerumque ’ As Dr. Stapf very justly points out, it is not possible to decide the identity of my Form ii. with Meyer’ s racemosa until we are in a position to compare Meyer’s orginal specimen of this species. It seems advisable, then, for the present to refer my Form iii. tentatively to Rhizophora racemosa, and to regard my Form ii. ew species, for whi ch I propose the name Rhizophora Harrisoni.t I suggest the following diagnosis of my Form iit. _ Rhizophora racemosa, G. FY. W. Meyer (descr. emend.); a k. Mangle, Dis inflorescentia dichotomo-paniculata, a a R. Harri- sonit, Leechm 2 See” crassissimo et radicula longissima ad 65 em. oes distinc rbor magna, ad ae m. alta, radicibus aeriis a ramis superioribus descendentibus. Folia 10-12 em. longa, 4-5 cm. ata, supra saturate virentia, —t, punctis nigricantibus adspersa. Petioli crassi, 1-5-3 longi. mi et ramuli inflorescentiae crassi, joke teil ickes-sbeddacull plus in 3 em. articuli 10-5 longi—in utraque dichotomia ° -margine calyciformi sexdentato enati. issima, qui s vel sexies repetito-dichotoma, articulata, mations, ‘Mabastra pallide flavo-viridia, ovata vel elliptica, 10-12 x 6-7 Calye viridis vel externe pallide flavescens, profunde anys artitus persistens, segmentis crassis ovato- lanceolatis subacuminatis coriaceis 10-12 x 6-7 mm. leviter patentibus. Petala 4, candida, linearia, Se margine obsolete hirsuta, inter lacinias calycis v Antherae saepe corolla paulo breviores, caducae. Suyhis: fortiter sulcatus; stigmata 2-3, rarius 4. Pericarpium 4-5 cm. longum, 2 em. atum, ovatum, crassum, scabriusculum, aliquando obsolete maculatum, saepe eurvatum. adicula omnium longissima, ad 65 cm. longa, pallide viridis, saepe curvata, apicem versus ferruginescens et sensim saepe ad 1°5 cm. incrassata. This aon abpent? to be the prevailing rive Bartica mangrove ae , and at iles up the Essequibo River, cd is the dominant * “Flora of Tropical sree vol. ii, p. 408. + In honour of Professor J. B. is C.M.G., M.A., Director of and Agriculture in British Guia: - rector of Science 3. calyx and pistil, 8. calyx and pistil, flower, M. Smi by Miss 3 Harrisonit.—1. bud, 2. flower, 5. stamen. racemosa (?)—6. 10. stamen. A arg te izophora 9. petal, Fig. Fig. B. 8 Form ii. may be defined as a new species, thus :— _ Rhizophora Harrisonii, Leechman; species nova a R. Mangle, L., inflorescentia composita, a Kk. racemosa, Meyer, pericarpio gracili et radicula brevi distincta. Arbor ad 20 m. alta, radicibus aeriis crebris a ramis superior- ibus descendentibus. Folia iis R. Mangle similia sed ferme elliptica, 11-15 cm. longa, 4-7 cm. lata, supra virentia, subtus punctis nigricantibus adspersa. Inflorescentia ramosissima, quinquies vel sexies repetito-dichotoma, articulata, multiflora. Rama et ramuli inflorescentiae graciles, pedunculi plus minusve 45 mm. longi, articuli seriatim breviores—20, 11, 9 et 7.5 mm. longi—in utraque dichotomia e margine calyciformi inaequa- liter sexdentato enati. Flores iis R. Mangle similes sed minores, 1°3 cm. longi, suaveolentes, praesertim sub occasum solis, demum campanulati. Alabastra flavescentia, ovato- lanceolata, 11 x 5 cm.. Calyx flavescens, profunde quadri- partitus, persistens, segmentis coriaceis lanceolatis acuminatis 13 x 4mm. demum late patentibus. Letala 4, candida, lineari- lanceolata, acuminata, margine saepe hirsutissima, demum inter segmenta calycis valde reflexa, quibus paulo breviora, caduca. Antherae 8, corolla breviores, aequales, caducae. Stylus leviter sulcatus; stigmata semper duo. Pericarpium rectum, plus minusve 4 x 1:5 cm. ovato-lanceolatum, scabrum et saepe maculatum. Radicula ad 30 cm. longa, omnino ferru- ginescens, The robust habit, dark-green foliage, stout panicles and huge radicles of my form iil. (? R. racemosa) make it very easy to distinguish it from R. Harrisoni even from a distance. I hope soon to be in a position to add details of the distribution in British Guiana of these three species of Rhizophora. I am indebted to Dr. Stapf for valuable advice and assistance, and I am also obliged to Mr. J. Rodway, F.L.S., of George- town, for the loan of the works of Jacquin and of Meyer to which reference has been made. III.—DISEASES OF PARSNIPS. STUDIES FROM THE PaTHonocicaL Laporarory: VI. A. D. Corton. (With Plates.) ke the matter the subject of special enquiry. Two visits for this purpose were paid to Evesham and district, and consider- able light on the subject has been obtained. At the same time ps, three of which were new to the country and one apparently capable of causing severe 9 damage, were noted. In the present paper the results of the investigations on Canker are recorded, as well as observations on the other diseases. CANKER. That a certain number of parsnip roots in all parts of England are annually disfigured through surface-injuries on the shoulder is common knowledge, and that in some cases serious decay follows is also well known. In certain localities this decay, which occurs in late summer and autumn, is very prevalent, and is termed “Canker”’ or “‘ Rust.’’ The name is loosely used and the trouble not easy to define; it has apparently never been specially investigated. The brownish-coloured rot shown by affected specimens is not a characteristic symptom, but is common to other types of injury, being due largely to the oxidation of substances in the mucilage present in the parsnip root. The term ‘‘ canker’’ as used by growers implies a-more or less open wound, at first brownish-red in colour, which affects the shoulde or upper part of the root, and frequently leads in bad cases to the destruction of the entire root (see plate iii). Damage in 1913 crop.—tThe trouble reported from Worcester- shire at first seemed to be of a more definite nature, the roots but was not found to agree satisfactorily with any described species of that genus. It was subsequently ascertained that this black rot was a special form of decay, and that canker was not associated with any particular fungus, but arose from a surface wound, which, invaded by various kinds of micro-organtsms, rapidly led to the destruction of the root-tissues. In the Evesham district it was estimated that 10 per cent. of the entire crop was lost through canker, and this, judging from figures supplied by Messrs. Yates, would amount to a loss of 370 tons of roots from seed supplied by their Evesham branch alone. In certain fields the loss was very severe; 50 per cent. of the crop in more than one case was stated to have been ren- dered unsaleable. é 10 largely bacterial, the initial injury in most cases being probably ue to some other cause. No specimens of the black Phoma rot were seen. It was ascertained that a more or less severe form of canker occurred over a wide area around Evesham, also in parts of Gloucestershire, whilst roots received from Somerset were almost completely destroyed through the same cause. In some cases where no rot, either wet or dry, was present a certain amount of the so-called canker was definitely attributed to the workings of Carrot Fly, and injury by this pest was found subse- quently to be very considerable (see Bd. Agric. Leaflet No. 38). . At this season it was obviously too late to obtain a general insight into the nature of canker and of the black rot. A few simple tests, with regard to soil and culture for the coming season, were therefore arranged pending further examination in September, about which date canker was said to commence. These experiments included the dressing of the soil with lime, soot and sulphate of ammonia. is was carried out by several 0 Evesham. The fields of seven large growers were inspected, and also the gardens of several others where parsnips were grown on a smaller seale. By this date canker had been reported from several market gardens, and, as was subsequently discovered, was in certain localities far advanced. It was regarded as being To form a really true estimate it would have been necessary to lift a large number of roots and have a somewhat detailed knowledge of the character of the soil and its previous history with regard to manuring. This much, however, may be said: in most fields the distribution of the canker appeared fairly uniform and to be unaffected by the dressings which had been apphed. The application of sulphate of ammonia did not apparently result in any very great increase in the growth of the parsnip nor in the amount of disease. No very striking results were, however, anywhere apparent, though, on the whole, canker appeared to be distinctly worse where limine in the past had been omitted. cases on wet heavy land (Pebworth and Sedgeberrow). A bad case was noted on light land (Fladbury), On the whole, canker appeared least prevalent on medium loam. 11 Animal and fungus injury were next searched for. The former was observed in some variety. The workings of the carrot fly larvae were manifest both in field and garden crops. Slug-injury was also noted, and the occasional presence of wireworms and surface caterpillars (Agrotidae), locally termed “‘ leather grubs,” doubtless accounted for other damage. The millipede (Julus pul- chellus) was often abundant in the decaying parts. se animals, however, were far too few in number and too irregular in distribution to be for a moment considered as being primarily re- sponsible for canker. Carrot fly was the most general, but it was usually sparsely distributed except in small gardens and allot- ments, and the characteristic tunnellings which the larvae pro- duce, not ouly on the shoulder but in all parts of the outer tissue of the root, form a very definite and distinct type of injury. Cause of Canker.—The cause of canker was more satisfactorily ceasing to search for all indirect and subsidiary causes, wa e the subject of special investigation during the remainder of the visit. ‘The cracks were, as a rule, } to 2 in. long a 5-3 in. The fresh wounds were perfectly clean (Plate iv, figs. 1 and 2), but the cortical tissue within soon became attacked by animals, such as slugs and centipedes, and by various fungi and other The cracks were obviously due to the expan- as a result of other wounds, such as carrot-fly injury is possible and even probable, but it was obvious that a very large propor- tion of the trouble in Worcestershire is, primarily, due to this the periderm. The cause of crack-formation and the inner tissues is discussed later. canker originated from these surface lesions. That it may follow rupture of subsequent decay of Experiments at Kew.—Although the origin of the canker had been traced to the surface lesions described above, the organisms ‘eoncerned with the actual decay had not been classified and 12 isolated. ‘Their distribution in the soil and their relative abund- ance in different localities was unknown. For this reason the results of the soil ha at Kew arranged the previous spring may be recor ere. Five aad of Parsnips were sown in April 1917, and were treated as follow Plot A inoculated with 2 bushels of infected Evesham soil. Plot C inoculated with cut-up diseased roots from Evesham. Plot D inoculated with cut-up diseased roots from Somerset. Plots B and E served as controls. The diseased roots and soil which were obtained from Worces- tershire and Somerset in March were added with a view to inocu- ating the soil with any organisms, parasitic or otherwise, which might be concerned with the production of canker. The variety of Parsnip sown was the same as that very largely used in es parece a broad-shouldered form of ‘‘ Hollow Crown.” Preliminary ‘examination in August and September showed the See of many small cracks, but no marked development of caer The plots were finally cleared on November 12, by ich date a Sedetdarahte number of cankerous wounds ha develiined: The results are set out in the accompanying table :— | \ per ; cent. Pot A.—Inoculated with Evesham | Sound... ... ...| 65 | 68 soil (total number of plants = 96). | Cracked — Reoby 18 _Cankered or r commencing 14 14 to canker Piotr co eae (total number of! Sound _... “a pe a 3 58 plan 178). | Cracked 42 23 C ankered or commencing. 33 19 to canker ners C.—Inoculated with iscened Sound __... se ee OU 66 oots from n Kvesham (totalnumber | Cracked 3 17 of Eakin = 76). | Can nkered o1 or aed 13 17 | a D.—Inoculated with diseased| Sound... ane eis Oe 63 from Somerset (total number | | Cracked . 22 22 oF plants = 99). | Cankered or r commencing | 15 15 te nker Piotr ter (total number of | Sound __.., La Pip tP ee 70 plants — 168 | Cracked . 26 15 | Cankered 0 or r commencing 24 15 to canker From the above it will be seen that from 15-23 per cent. of the cracking occurred in all plots, and from [4-17 per. cent. of canker. he figures for the latter are high, but except in plot C the canker was in an early stage, and the root were marketable and not a apeerciably damaged. In plot C eight roots had bere decayed, but from the presence of equally severe injury in some neighbouring plots (not included in the experiment) this 13 cannot, without further proof, be attributed to the effect of the inoculation with diseased roots. It may be due to early ans and early commencement of decay. The experiment tends show that inoculation with diseased material had no effect, since of the controls one (B) showed the maximum amount both of cracking and of canker, and the other (E) almost the minimum.* n addition to the above, efforts were made by means of direct checirvatien of\ roots in the Kew plots to follow the course of development of the cracks and the subsequent decay. The results were not conclusive—a fact which may be partly due to the obser- vations being made late in the season, namelv. in September, after the return from Worcestershire. Nine prowthaipwtle on roots in plot B were carefully examined and measured, the roots being left in situ, the soil merely scra tee away for purposes of inspection, and replaced at once. The weather was somewhat dry for the first ten days after ener ee After two weeks the plots were examined, when but little extension of the cracks and no decay had taken place. Two weeks later, after considerable rain, a little decay had commenced and some new cracks appeared. After another fortnight, i.e., six weeks after mark- ing, the roots were lifted and examined, with the following results:—Three slits had not increased in size and remained sound, 3 had distinctly increased in size yet remained sound, 6 or 7 new cracks had developed, all of which remained sound. In another set of roots on the same plot a series of shallow vertical cuts on the top of the crown resembling growth-cracks were made with a knife. Examined 6 weeks later, four remained sound and two developed into a cankerous wound. A similar experiment was performed on plot C (infected soil) with the addition of 12 horizontal cuts on the side of the root. All these cuts, however, remained sound. It was noted that an pa surface attracted slugs, = ae -injury being obvious both n growth-cracks and artificial cuts Summing up the Kew experiments, it may be said (1) that growth-cracks occurred at Kew as in Worcestershire, but they were decidedly smaller in size and later in OO re (2) that canker followed in a few cases as a result of these or other injuries; and (3) that canker was found to is no more frequent the plots inoculated with Evesham soil or with diseased roots from Evesham than in the ordinary Kew soil. The less ae Tt may perhaps be accounted for by the small size e his was due to the seed being sown late (nid April) saa the soil being a heavy compact silt which had not been manured for two years. * Experiments carried out by Messrs. ‘cee: at poise for testing the effect of manures showed a distinctly larger perce! er than at ew. The use of lime proved highly penie aeses Pboth an eae to size of roots and reduction of canker. e use of manures me, such as Pernvian guano, s of ammonia and pease of sreatgee not successful in mitigating the disease. 14 Anatomical considerations.—The parsnip root consists of a central core of xylem surrounded by a mass of soft spongy paren- chyma. The whole is bounded by a thin periderm., The rela- tive amount of core and soft outer flesh differs somewhat in dif- ferent varieties, but as a rule the fleshy tissue is largely in excess of the core, and it is the aim in selection to keep the central core as small as possible. The core is composed almost exclu- sively of secondary xylem, but it contains a very large amount of xylem parenchyma, hence it is comparatively soft. The outer fleshy portion consists essentially of secondary cortex; it is com- posed entirely of parenchymatous and other soft elements which are produced abundantly from the cambium. The cells are ensely packed with starch. Through the cortex runs a system of mucilage canals, which are especially found immediately beneath the periderm. The secretion may be seen exuding from roots on freshly cut surfaces as a white milky fluid, which turns reddish-brown on exposure to the air: this change of colour, as is shown by the guaiacum test, is due to the oxidation of some substances contained in the fluid. . The periderm surrounding the root is formed from a phellogen which arises in the inner tissues of the root, probably in the peri- cycle, at a very early stage. In the young growing apex of the root the phellogeh and three or four layers of periderm are clearly visible in transverse section. In older roots the periderm appears, owing to tangential stretching, to be less regular. In median longitudinal section, however, the regular arrangement of the cells, obviously of secondary origin, is quite clear. This thin layer of periderm is interrupted only in the furrows mark- ing the emergence of the lateral roots, where a thick pad of corky tissue is present. It is obvious to the naked eye that the cortical tissue exposed to the air as a result of cracking does not heal over. Anatomical examination shows, however, that an attempt is made on the part of the plant to protect itself, inasmuch as the outer and side walls of the exposed cells become suberised. A few cell- divisions also take place, but there is no trace of a definite phellogen or of a layer of wound-cork. This protection is quite inadequate to keep out micro-organisms, the inter-cellular spaces and cracks or fissures due to drying providing ample means of entry. — The soft parenchymatous tissues are, moreover, specially attractive to slugs, an immense amount of slug-injury being noticeable if large gaping slits are made with a knife. It is therefore not surprising that decay sets in and canker follows, and the cause may be said to be entirely due to the inability of the parsnip to form a protecting layer of wound-cork. The deep open clefts which occur in the roots of swedes, carrots, mangolds, and turnips become protected on their surfaces by a thin cork layer. No cork layer is formed when such deep cuts are made artificially in the parsnip. It would appear, therefore, that the hin ance to cork-formation is not due to the different method of vie ein but to some physiological peculiarity of the parsnip 4 15 The Causes of Cracking and Remedies.—The formation of growth-cracks is a phenomenon familiar enough to gardeners, and is of common occurrence not only in root crops, but in fruits such as apple, pear, and cucumber. In the pulpy fruits of plum, cherry, and tomato a cracking of the skin is also frequent, and -in this case, as in the parsnip, decay results. Cracking is brought about by an unequal growth in the struc- tures concerned; the inner tissues grow more rapidly, and ulti- mately rupture the outer layers. In the case of the pulpy fruits mentioned, cracking may be partly due to excessive turgidity of the inner cells. In either case it is, as a rule, due to the effect of,an abundant supply of water at the root after a dry period, but at times it appears to be merely the result of rapid growth, _when the expansion of the surface tissues does not keep pace with that of the inner. The effect of hot sun and dry conditions, y, if there is not a copious supply of water at the root, is to ripen growth, and this entails, amongst other things, loss of elasticity in the cell-walls and decrease of the power of cell- division. Dry conditions, whether in the case of fruits or roots, are felt first at the surface, and hence surface tissues are more subject to premature ripening. f late summer rains stimulate new growth, the outer, partly ripened tissues respond more slowly, and if the pressure from within be excessive they are liable to become ruptured. The above considerations apply to the parsnip and explain to a certain extent the cause of cracking. It is obvious that the very wet September of 1916 following the warm, dry August was conducive to late and rapid growth, which would, for the reasons given, lead to extensive cracking of the surface tissues of ~ the root, and an unusual abundance of canker. In 1917 canker collected from various growers cracking was general in 1916 in parsnip-growing localities though the losses elsewhere were much less serious than at Evesham. One may inquire therefore as to the presence of special conditions at Evesham, which might tend probably a tendency to premature ripening. This theory receives support from the fact that several growers maintain that liability to canker is lessened by late sowing. The soil itself may be too rich. Though fertilizers are never applied directly to the crop, manured, an evidence, in the case of certain growers at all events, that parsnips are grown on the same land for two years in succession and that — a proper rotation is not given. The effect of the latter course is not only unfavourable for parsnip growth, but s to stock 16 the soil to the full with the particular organisms, both animal and vegetable, concerned in their decay, and hence if cracking occurs the parsnip might be expected to be rapidly destroyed. Lastly, liming has been neglected. Experiments carried out b Messrs. Yates in 1917 show a decided improvement where lime was applied. Yet another cause of the severity of the canker attack may lhe in the nature of the variety of parsnip grown. As already men- tioned breeders aim at a small core and ample flesh, an ideal well seen in the Evesham parsnip root, which is noted for its bulky fleshy top. Such a soft mass of parenchymatous tissue would probably be easily stimulated to extensive growth after rainy periods during the summer months and to rapid expanston of the inner tissues. The breeders may have exceeded the limits in that direction and produced a parsnip particularly subject to a physiological defect. As evidence in favour of this view 1t may be mentioned that in allotments in Worcester, though the long slender parsnip such as ‘‘Student’’ showed rupturing of the rind and a certain amount of canker, these were not so ‘extensive as in the Evesham parsnip. Preventive Treatment.—In order to reduce the amount of surface-cracking and canker the following treatment, based on the above conclusions, may be recommended : — 1. Too rich a soil must be avoided. 2. Late sowing should be adopted. Such observations as it has been possible to make in 1917 tend to confirm growers’ statements that plants sown in the end of April or beginning of May suffer much less than those sown in February. 3. Liming must not be neglected. As well as improving the tilth, lime acts by liberating reserves of nitrogen and potash, and its effect on the parsnip is seen in the im- proved quality of the crop both in size of roots and decrease in the amount of decay. 4. A dressing of salt has been found very effective by some — growers. On heavy soils 5 cwt. per acre should be applied, and on light soils up to 10 ewt. per acre may be used. The salt may act by liberating a certain amount of potash from the soil, and one of the most marked effects of potash is to retard maturation and to enable the plant to continue its vegetative growth. It is possible that this effect may extend to the phellogen (rind-producing layer) and consequently render it less liable to rupture. For the same reason potash manures should prove beneficial. . 5. A proper rotation should be given, and parsnips must rors be grown for two years in succession on the same Summary.—The canker of parsni : p or the decay of the upper part of the root during late summer and aiken is a diease 17 well-known. to growers, but one which has apparently never been investigated. In certain districts it has recently been on the increase and is responsible for serious losses. It is shown to be due primarily to a physiological phenomenon which causes the surface tissues to become ruptured, and not to the invasion oi a fungus parasite. Cracking takes place during the growing season especially if rains follow a dry period, the portion involved being the skin, i.e., the periderm and the outermost ‘layers of the cortex. The cracks, which for the most part run horizontally around the upper part of the root but stso in a vertical direction, are from 4 to 2 in. long and gape open exposing the soft inner tissues. The “‘ canker ’’ or decay which follows is shown to be the result of the inability of the pear to form a layer of wound cork. Though the outer walls of the become suberised and few cell-divisions occur no definite formation may be favoured by certain methods a view to retarding maturation. Orner Parsnip DISEASES. Erysiphe Polygoni, DC. The Parsnip Mildew. This fungus, which attacks many cultivated plants, occurs frequently on Parsnips, being found very plentifully in the market gardens around Evesham and Wisbech; it has also been noted elsewhere. The perithecia are found late in the season both on the leaf-blade and the petiole, but especially the latter. The attack is not sufficiently severe to necessitate treatment. Phyllachora Pastinacae, Rostr. Plantepatologie, p. 511. New to Britain. e conidial (Cylindrosporium) stage of this species was found attacking a certain number of parsnips with great severity in a market garden at Mickleton (Glos.). The disease shows itself first as brown spots on the upper B 18 surface of the leaves. These later coalesce and form extensive rown ereas. On the under side the minute sori are visible as black shining areas, just visible to the naked eye though they may ultimately reach 2 mm. in diameter. Occasionally they appear also on the upper surface. In bad cases the whole foliage including the petiole is attacked and killed, and as it becomes browned and scorched in appearance diseased plants stand out clearly in the field. ‘Although neither the ascigerous nor conidial stages of . Pastinacae have been previously recorded for this country there can be little doubt that the fungus occurs frequently, but has been overlooked. Further it is probable, since their morphological characters are identical, that P. Pastinacae is synonymous with P. Heraclei which is not uncommon in England on Cow Parsnip (Heraclewm Sphondylium). The ow widespread and injurious Celery Blight (Septoria Apii) origin- ally spread to Celery from the wild Apiwm (vide Pethybridge, Journ. Roy. Hort. Soc. vol. xl., pp. 476-480). The spore measurements of the conidial stage given in Saccardo are incorrect, and a revised description is given below.* P. Pastinacae, Rostr.—Cylindrosporium Pastinacae, Lind. Danish Fungi, p. 493. Septoria Pastinacae, Westend, Herb. Crypt. Belg. no. 639. Sori on the under side of the leaf, scattered or gregarious, not on distinct spots, immersed in the tissue of the leaf. Pyenidial wall lacking, spores set free through the ruptured epidermis. Conidia hyaline, strongly curved, falcate, ends rounded, at first with a few oil-drops, finally 1-septate, rarely with a second septum, 50-70 x 4-5 pn. Ramularia Pastinacae, Bubdék. In crowded gardens and allotments the spotting was severe and sufficient seriously to weaken the plants; in the open fields the damage was insignificant. The synonomy and description are as follows :— * For assistance with regard to the identification, synonomy and descrip- tion of these new records and al at . D so for the text figures, I am much indebted to Miss E. M. Wakefield of the Kew Gabesuer’ sce 19 R. Pastinacae, Bubdék in Sitzber. bbhm. Ges. Wiss. Prag. Sep. 1903, p. 19. ? R. Pastinacae, Lindr. et Vestergr. in Acta Soc. aun. Flor. Fenn. xxii., No. 3, 1902, p. 8 (non Cercosporella Pastinacae, Karst. in Hedwigia xxiii., 1884, p. 63). Fig. 1.—Ramularia Pastinacae. Portion of leaf showing angular spots (nat. size). (b) Tuft of conidiophores emerging through a stoma (x 850). (c) Spores (x 850). Spots on the leaves, definite, more or less angular, limited by the veinlets of the leaf, 1-2 mm. in diameter, at first dark greyish- green, then brown; spots on the petiole, dark, much elongated up to 2 cm. long 2-3 mm. wide. Sori minute, crowded, chiefly on the under-surface, but also epiphyllous, white or very pale rose-tinted. Conidiophores fasciculate, emerging from the stomata, cylindrical, unbranched, finally more or less toothed towards the apex, 30-75 x 2-4-4 u. Conidia cylindrical, straight or very slightly curved, ends rounded, at length 1-septate, or rarely 2-septate, sometimes in short chains, 15-27 (9-35) x 3-5-4 Text figure la, b, c. Cercosporella Pastinacae, Karst. ew to Britain. Found on leaves and petioles and closely associated with Ramularia Pastinacae. Only a slight attempt _ was made to separate the two fungi in the field from which it appeared that the Cercosporella is by far the least common. When unmixed with Ramularia the spots are round (text figure 2a), but as a rule the two fungi occurred together and their general distribution appears similar. With regard to the identity of the fungus there appears no doubt that it is the same as Karsten’s. Lindroth and Vestergren | ; B2 20 (Acta Soc. Faun. Flor. Fenn. xxii., 1902, No. 2, p. 8) stated that Karsten’s fungus was a Ramularia, with 2-celled, rod- _shaped conidia germinating soon after maturity. This descrip- tion would apply to Ramularia Pastinacae, Bubak, but not to Karsten’s own description of the long, tapering conidia. In view of the close association of the two species noted in England, it seems legitimate to conclude that Lindroth and Vestergren must have seen only the Ramularia, and erroneously supposed it to be Karsten’s Cercosporella. They do not, moreover, state whether they examined the type material. The description is as follows :— Fig. 2.—Cercosporella Pastinacae. (a) Portion of leaf showing round spots (nat. size). (b) Tuft of conidiophores with young spores (x 850). (c) Spores (x 850). C. Pastinacae, Karst. in Hedwigia, xxiii., 1884, p. 63. Spots rounded, not bounded by the veinlets, 1-3 mm. in a qn miophone short, ) #. Conidia hyaline, slender tapering craduall ; i Et : bg g ‘a mas Botlade the apex, with 3-6 indistinct septa, Text figure 2 a, b, ec. ‘Kew Bulletin, 1918. ILL. PARSNIP CANKER, rep. ait 9 [To face page =v. Kew Bulletin, 1918. | LV: PARSNIP CANKER. To face page 21.) Plasmopara nivea, Schroe In 1917 this well-known antes on the U eo appeared on parsnips late in the season, _— noted abundantly near Wisbech in the end of September. In October Messrs. H. Wormald and E. W. Sow forwarded material from near Ashford and Haslemere respectively, and signs of its presence were observed later in several other — The fungus shows first as pale spots or ill-defined areas on the upper surfaces of the leaf with a delicate white mould peotmidiag below. With age the spots increase in size and become dark in colour, form- ing irregular blotches several centimetres across, similar in form and appearance to those caused by Phytophthora infestans in potato, a character which distinguishes it from any of the other diseases previously mentioned. In one garden near ‘Wisbech where parsnips were grown between fruit trees, a considerable amount of the foliage was destroyed by this fungus. Expranation oF Pirates IIT & IV. III. ‘Typical example of Parsnip canker. IV. Illustrating formation of growth-cracks and subsequent development of canker: (1) ee cracking; (2) hori- zontal and vertical cracking; (3) comm encement of decay ; (4)-(6) further stages of decay and pibducted of canker. IV.—TAGASASTE AND GACIA. (Cytisus spp.). J. Hurcuinson. Amongst a donation of Canary Islands’ seeds for distribu- Se tion, Dr. G. V. Perez, of Orotava, Tenerife, has forwarded to Kew several pounds of “ Pagasaste,” Cytisus proliferus, var. palmensis,* Christ, ‘‘Gacia,”’ Cytisus stenopetalus, Christ, and ** Gacia eee Cytisus pallidus, Poir. In various communi- cations to erez has repeatedly urged the importance of Canary Island fodder plants, especially the “ Tagasaste,” for planting in warm dry countries, such as North and South Africa, Festal tes and the near east; whilst ‘‘ Tagasaste’’ was the subject of a pamphlet by Dr. Perez’s father, Dr. V. Perez, and Dr. P. Sagot, published in’ Paris in 1892. The Kew Bulletin for 1891, pp. 239-244, contains a number of extracts from the Kew Reports giving particulars of the abt of ‘*Tagasaste ’’ seed from Kew to various Colonies, and some of the results therefrom; and Sir Daniel Morris, in the Kew Bilin, 1893, pp. 115-117, gave his own personal observations of the junction with the present notes. Recently Mr. J. H. Maiden has given a comprehensive account of the shrub in the Agricul- Gazette of New South Wales, for October, 1915, pp- a he mya the present paper treated as a separate species, C. _palmensis, Hutchinson. Re 883-887. In the same journal (p. 888), there is a government rowing ‘* Fodder Vent.), “‘Mulga’’ (Acacia aneura, F. V. Muell.), “ Jarob ean’’ (Ceratonia siliqua, Linn.), ‘‘ Tree Lucerne”’ (Sida rhombifolia, Linn.), and the native ‘‘Salt bush” (Atriplex campanulata, Benth.). From what we know of ‘“ Tagasaste, it seems probable that it and the other two Cytisi mentioned above might be suitable for inclusion in such a list, as they appear to be particularly adaptable for this purpose. It is true that in some places, notably South Africa* ‘“Tagasaste the shrub is an acquired one, and they must first of all be taught to eat it, if necessary, by a little starvation. The seeds are as hedges and clipped annually for use during drought when s : As the result of a critical examination of these Canary Island Cytisi, it has been found that three distinct species have been confused under the name of C. proliferus, Linn. They are dis- tinguished from one another not only by striking differences in the shape of the leaves, and expecially y their hairy covering, but ack is characteristic of either one or at most two islands of the Canary Archipelago, whilst there is, in addition, a differ- ence in value as fodder as well as constant segregation under he peasant people of the Canaries, out their crops and plants. e extreme endemism of the species of Cytisus in separate islands is remarkable, two of the islands, La Palma and Grand Canary, having each two species, and Gomera and Tenerife one i i hem. is subject, and a detailed genus, may, however, be } paper. e following is a summ of the most obvious differences ary of among the three forage plants mentioned above and of the species with which they have been confused. Se REMAND NE ete ttc pale * See the South African Agricultural Journal, v. 134 (1913). 23 *Leaves densely silky-hairy on both surfaces :— 1. Cytisus pallidus, Poir.—Laxly leafy; leaflets rather elongate-lanceolate, averaging about 4 cm, long an 7 mm. broad, silky-hairy above, silvery below with long closely appressed hairs; bracteoles as long as or longer than the calyx; calyx-lobes linear-subulate. ndemic to La Palma. Native name ‘“‘Gacia blanca.”’ 2. Cytisus Perezii, Hutchinson, sp. nov.—Very densely leafy; leaflets distinctly obovate, averaging about 3 cm. long and 1:2 cm. broad, densely clothed with long closely packed silky hairs on both surfaces; brac- teoles much shorter than the calyx; calyx-lobes tri- angular-lanceolate. Endemic to the islands of Grand Canary and Hierro. Native name ‘“‘ Escabon de Canaria.” **D caves densely silky-hairy only on the lower surface, the upper surface glabrous or nearly so :— 3. Cytisus proliferus, Linn.—Rather densely leafy; leaf- lets linear or linear-oblanceolate, averaging ~ about 3 cm. long and 0-5 em. broad, glabrous or very nearly so above from the beginning, but especially so when mature, closely silky-hairy below. Endemic to Tenerife and Gomera. Native name ‘* Escabon.”’ *** eaves glabrous or thinly pubescent with few short scattered hairs :— lanceolate. Endemic to La Palma. Native name ‘‘ Tagasaste.” 5. Cytisus stenopetalus,* Christ.—Laxly leafy; leaflets obovate or obovate-oblanceolate, averaging 25 cm. long and 0-8 em. broad, thinly and shortly pubescent; racemes overtopping the leaves, man -flowered ; brac- teoles half as long as the calyx; calyx-lobes triangular- lanceolate. Endemic to La Palma, Gomera and Hierro. Native name ‘‘ Gacia. PS * 1. Cytisus pallidus, Poir. Encycl. Suppl. ii. 442 (1811); Bot. Mag. t. 8578. ee ee eee ee Bee ee ee * Readily distinguished from C. maderensis, Masf., from Madeira, with which it was united 24 Genista splendens, Webb. in Webb. et Berth. Phyt. Canar. ii. 43. (1836-50). Teline linifolia, var. latifolia, Webb. lc. 42; Pitard et Proust, Fl. Canar. 151 (1908). Cytisus linifolius, var. pallidus, Briq. Les Cytis. des Alpes Marit. 140 (1894). Canazy Isuanps. Palma: Barranco del Rio, 1843, Webb; cliffs above Tenerra, Caldera, Sprague and Hutchinson 466; cul- tivated by Dr. G. V. Perez at the Villa Orotava, Tenerife. This is reputed to be a valuable forage crop (vide Dr. Perez), and is known to the peasants of La Palma as “‘ Gacia blanca ”’ ; the local name in Webb’s time was “‘ Herdanera.’’ It is, how- ever, neither so good nor so vigorous a forage plant as ‘‘ Gacia,”’ C. stenopetalus. 2. Cytisus Perezii, Hutchinson, sp. nov.; C. proliferus, var. Canarie, Christ in Engl. Bot. Jahrb. ix. 120 (1887) var. lamiflorus, O. Kuntze, Revis. Gen. Plant. i. 178 (1891). Frutea vel arbor parva, multe ramosa; rami breves, subdense foliati, dense molliter tomentosi. Folia brevissime petiolata, tri- foliolata; foliola obovata vel oblanceolata-obovata, ad apicem breviter mucronatum rotundata, ad basin cuneata, 2-3°5 cm. longa, 1-15 cm. lata, tenuiter chartacea, utrinque pilis nitidis sericeis appressis permanente dense obtecta : petioli usque ad 1 em. longi, molliter tomentosi. Flores in ramulos breves axillares con- ferti; pedicelli usque ad 1°5 em. longi, nutantes, molliter fulvo- tomentosi. Calyz circiter 13 em. longus, extra dense fulvo-vil- losus. Veawillum orbiculare, breviter unguiculatum, circiter 1°8 cm. longum et 1°3 em. latum, apice leviter emarginatum, dorso sericeo-pilosum; alae oblique oblongae vel semilunares. apice rotundatae, uno latere undulate auriculatae, ungue 5 mm. longo glabro, lamina 1°5 cm. longa latere inferiore breviter ciliata ; carinae petala aliis simillima sed minora et longiore auriculata. Pubus staminalis 1-3 em. longus. Ovarium appresse villosum, stylo curvato basin versus parce pubescente. Fructus non visus. Canary Istanps. Grand Canary: Firgas, 0. Kuntze; without definite locality, Bourgeau; Lowe: Finlay. Hierro: About a mile from Valverde, towards the port, Lowe 70; aboye the church at El Golfo, Lowe 70. Vernacular ‘‘ Escabon de Canaria.”’ It gives the writer much pleasure to associate with this species the name of Dr. G. V. Perez, who has provided a liberal supply of his cultivated Canary Islands’ Cytisi, and whose kindness to the writer and his companion during their visit to Tenerife will not be readily forgotten. 3. Cytisus proliferus, Linn, f. Suppl. 328 (1781). liferus, vars. angustifolius et nanus, O. Kuntze, Plant. i. 178 (1891). Canary Istanps. Gomera: El Monte, Hermigua, Lowe 128. Tenerife: El Monte Perexil, above Agua Mansa, 2000 m., Lowe 9; below Pedro-Gil, 1600 m., tree 20-25 ft. on bate slopes, Dinn 157; Santiago, O. Kuntze; Barranco de Montijo, 1000-2000 m., Mann 2514; ‘‘ Mountain woods,’’ Masson (+ ideals Talbot; Bourgeau 51; 1306. oe C. pro- Revis. Gen. 25 Vernacular ‘‘ Escabon.” e ty ecimen was collected in 1778 by Francis Masson, the first collector sent out by Kew, and is at the Natural History Museum, South Kensington. 4, Cytisus palmensis, Hutchinson, sp. nov.; C. proliferus, var. palmensis, Christ in Engl. Bot. Jahrb. ix. 120 (1887). ruter laxe ramosus, usque ad 10 m. altus; rami elongati, brevissime pubescentia ; petioli 1-1-5 cm. longi, basin versus com- planati, molliter tomentosi. Flores albi, in ramos breves axil- lares apice foliatos 3-4-natos dispositi; pedicelli graciles, 1-1-5 cm. longi, pilis longis et brevibus dense tomentosi. Calys cir- citer 1 cm. longus, extra molliter pubescens, labio superiore breviter et acute bilobato, inferiore parum longiore acuto. Vewillum orbiculari-obovatum, breviter unguiculatum, fere 2 cm. longum et 1-4 cm. latum, apice minute emarginatum, dorso rotundatae, uno latere auriculatae, ungue ciliato 5 mm. longo, lamina 1-5 cm. longa basin versus minute ciliata; earinae petala aliis simillima sed paullo breviora. ubus staminalis 1 cm. longus. Ovarium pilosum, stylo curvato parce pubescente. Fructus oblongus, 4-4-5 cm. longus, 1-1-3 cm. latus, subappresse pubescens. Semina oblongo-ellipsoidea, 5 mm. longa, nigra, nitida, basi arillo crasso cupulare 1°75 mm. alto crenulato cir- cumdata. Canary Isnanps. La Palma: Borders of woods above the Barranco del Carmen, Christ; Dolores, Christ; La Banda, above Argual, at the foot of the cumbre towards the Caldera, Lowe; Lomo de Tagasaste, Caldera, Sprague and Hutchinson 436. Vernacular ‘‘ Tagasaste.”’ This plant is evidently confined in a wild state to the island of Palma, where it is extensively cultivated as a fodder plant (vide supra). 5. C. stenopetalus, Christ in Engl. Bot. Jahrb. ix. 162 (1887); Pitard et Proust, Fl. Canar. 151. T'eline stenopetala, Webb. in Webb et Berth. Phyt. Canar. ii. 37 (1836-50). Genista stenopetala, Webb. I.c. t. 45. Telinaria stenopetala, Presl. Bot. Bemerk. 49 (1844). Cytisus maderensis, Briq. Les Cytis. des Alpes Marit. 137, partim (1894), non Masf. Canary Istanps. La Palma: Barranco del Rio, Webb; in the woods at El Monte de Barlovento, May, 1858, Lowe 342; between Garafia and Barlovento, Feb., 1888, O. Kuntze. Gomera: El Monte, Hermigua, Apr. 1861, Zowe 117. Hierro: Monte de Savinosa, June, 1845, Bourgeau 90; El Golfo, May 1855, Perraudiere; Lowe 202. , Vernacular ‘‘ Gacia.”’ : Briquet (l.c.) says this species occurs in the Cape Verde Islands, but the writer has not seen any specimens from there. 26 VI.—SPARTINA AND COAST EROSION.* Ipa M. Rorer. An experiment is being tried in North Somerset to fight against the constant erosion of the coast line in the Bristol Channel where the conditions render it entirely different from what has been undertaken hitherto in other parts. Here the spring tides rise regularly to 36 ft., and deposit on each occasion a considerable amount of mud right up to the limit of high water, but, on the other hand, scour away equally large quantities when the waves beat on the coast under the pressure of winter gales. The Bristol Channel is full of long stretches of mud flats with deep water channels running amongst them. As the tides ebb and flow the edges of these flats are continually changed, and this apphes equally to the long slopes of mud which are exposed at low water on the actual shore of the Channel. When these shore flats become lowered by the removal of their surface, the up-rushing tides, moving at times at five knots an hour over the round, reach with increased force the earth-banks, which stand across the Channel between Cardiff and Weston-super-Mare, but in other parts of the same stretch the destruction has been mark. In this way they prevented specially high tides in com- bination with strong gales from washing away the surface of the ourse of yea both higher up the Severn and lower down as far as Bridgwater, at those parts where such sea-banks could give protection to the This system of guarding the land answered very well, although a certain amount of ground between the sea-bank and high-water _ mark was being constantly eaten away by exceptionally heavy Gard. Chron., ser. 3, xliii. (1908), p. 33 i oe Sey: (i914), > 76-89 pl iB , and in Proceed. Bournemouth Nat. Sherring in Proceed. Bournemouth Nat. Sc. Soe. iv. (1913), p. 49, pl. i, ii. 27 weather. But that had been foreseen, and the land being of little value for feeding purposes, because of the large quantity of salt contained in the soil, the owners allowed it to go on un- checked. In other parts of the kingdom, however, the losses of valuable land from similar causes had become so serious that enquire into the facts and to suggest remedies. t was then urged that much good might be done to lessen, and possibly to either the sand or mud and prevent erosion by the waves, and later on, b e ir g of the ground gradually out of the reach of all such scouring action. Spartina was among the grasses recommended to be planted as specially suitable where soft mud forms the only protection against erosion for the adjoining earth-banks; and as these are the conditions met with on the shores of the Severn in North Somer- set, the planting of this particular grass became of special interest to the landowners of the neighbourhood. Spartina Townsendi, the hybrid between S. stricta and alterniflora, which now covers thousands of acres of mud-flats in Southampton Water and Poole Harbour, is the one which has been planted in North Somerset. It is a vigorous stiff grass, reaching usually 2} ft. in height, with rigid leaves, long and pointed, standing off at an angle of about 60°. These rigid out- standing leaves are characteristic features, and easily distinguish it from the other two kinds. The stems and leaves are coloured a bright green when young, but turn in autumn to a brown hue with a beautiful golden sheen resembling ripe corn. On top of the solitary stem are the flowers in a stiff, branched spike grow- ing at one side only of each branchlet. The spreading of Spar- tina to new areas and distant parts is mainly due to the dispersal of seeds: but as soon as it has established itself in an area, the covering over of the area by the closing up of originally isolated patches proceeds much more by stolons than by the inter- polation of seedlings, chiefly because seeding is very irregular, 28 by 24 ft. of water at ordinary high tides, but its growth appears to be checked nearly completely if the depth exceeds 3 ft., whilst in shallower waters nearer the land it will put up with every hardship. Hitherto the growth of this particular species in England has been in quiet waters away from strong tides and the beating of heavy waves, and it may prove that such backwaters and river estuaries are its abiding home. It will thus be evident that a plentiful growth of Spartina T'ownsendi on a bare expanse of: soft mud daily ‘vastien by a moderate tide is capable of converting quickly such mud-flats into dry land and raising them above the level of ordinary tides. Should it prove capable of doing similar useful work in the presence of strong tides, a great help will have been obtained to check the constant washing away of muddy shores, and the growth of the grass would in course of time form a raised barrier protecting the adjoining earth-banks from erosion by the waves. This is the experiment that is being tried on the coast about three miles south-west of Clevedon, North Somerset, and there Spartina Townsendi has been deliberately planted to test whether it be possible to overcome the strong tidal currents of the Severn, and the very heavy waves which beat upon the shore whenever the prevalent south-west winds blow strongly. Special visits have been made to the spot this autumn to examine the results, and it is very interesting to find that the grass appears to have established itself with success, although the conditions are so different from those of the quiet estuaries of the English Channel. It may be well to mention a few particulars of the conditions prevailing on the land below Clevedon between the estuaries of the rivers Kenn and Yeo, and the accompanying map will make these clear. Inland there are fields (F), approached by a lane, protected by two lines of sea banks (O, I), the two barriers being built to make sure that if the front one is broken through by the force of the strong waves that beat in at that particular part, the water shall not be free to flood hundreds of acres of grassland that lie behind at only a few feet above the level of high-water mark. Outside the front sea-bank is a flat stretch of grassland, called saltings or salt-marshes (E), in parts raised some 20 ft. straight up above the mud, and in others much less: very narrow in their width, and constantly becoming narrower by the erosion of the t, it looks as if all this salting would have disappeared by this time had not the authorities put in _ Seaward there is a stretch of soft sti in width, extending to low-water mark, and the it would grow waves THAI “— RIVER KENNY 30 from further eating away the saltings, already so nearly destroyed. It is apparent at once that success is likely to attend the trial. There are sure signs of a vigorous and healthy growth of the grass, as the dots on the mud (S) show that it has already obtained a good foothold. The circular tufts are becoming much larger, and, whilst coming close together in some parts, are in others arranging themselves into miniature meadows across the run of the tide. The number of tufts amount to several thousands, and as the mud is continuous into the bank they can send up new plants by means of their far-reaching runners or stolons. It is notice- able that the outermost tufts run in a nearly straight line, and this probably marks the spot where the mud is covered at ordinary tides by 2 to 3 ft. of water, and left dry again after a short time of immersion. The distance of this line is from 59 to 100 yds. out from the sea-bank (O). At one part, the sea-bank on the shore opposite this stretch has been breached recently by heavy waves (B), and the path along it cannot at present be used. This arrangement of the Spartina tufts in lines parallel with the shore is very noticeable at present in three small bights, where the rush of tide may be slightly checked by the intervening head- lands, and a similar arrangement has been observed to occur in the quieter waters of the English Channel; but what is particu- larly interesting is the arrangement at the outermost belt, and it may be that the depth of mud rather than the height of the tide is the determining factor for the vigorous growth of the grass. Four years ago the landowner, Mr. A. E. Clothier, of Keyns- ham, obtained a quantity of Spartina Townsendi from Hayling Island, which he had split up into one thousand small tufts con- sisting of two or three plants in each, and these were planted by the Warden one yard apart in a straight line in the mud close against the upright earth-bank of the saltings. The waves, how- ever, promptly fioated them all out, and deposited them 40 or 50 yds. to seaward. There most of them fortunately took root, and gradually settled down in a more or less straight line opposite the place where they were originally planted, with many of them shifted nearly half-a-mile northwards. : It was further stated that in about the year 1900 there was over 6 ft. of mud on this shore lying upon a layer of peat 4 in. thick, under which is found a bed of white clay, but by 1917 the mud was only 6 in. deep on the same spots (T). It would seem that on such thin mud close into the bank it was not a good place to plant the Spartina, but it is remarkable and novel that the fufts should able to re-arrange themselves some way further out where the conditions were more favourable. : ‘o illustrate er the scouring away of the shore it was learnt that in the three bights mentioned the earth-banks had been faced with carefully placed stonework, which for some years 31 past had to be constantly extended in length, and more often had to be underpinned with additional stones. Only this past summer at one spot 1} ft. of underpinning had been put in, and within six weeks the operation had to be repeated, because the scour had taken away the supporting mud beneath for a depth of some additional 18 in. The little tufts have increased greatly in number and have grown remarkably in size, many of them, after four years’ growth, being more than one yard in diameter and forming many offshoots. Scarcely any of the tufts, however, have so far had time to grow sufficiently large actually to join together, and all of them in their present position are covered by a foot or two of water at each tide for one or two hours. If the present rate of growth continues with equal rapidity the mud will soon be clothed with a dense vegetation, and, to judge bv the inspection of one of the present tufts, the hitherto soft mud will afford suff- cient support for a man to stand upon the ground without making more than a footmark. The numerous stems keep back amongst them during each immersion a small quantity of mud and silt, which serves to raise the ground, a little being held by the roots and stems, and at the same time a certain amount of sand brought in with the tides is left on top, which still further solidifies the mud. Four years is a very short time to watch such changes, and it will be interesting to see whether Spartina can hold its own against the force of the waves, and increase in sufficient quantity to render their action of small account after they have travelled across the thick growth to the foot of the saltings. Should the c for that purpose. VI.—MISCELLANEOUS NOTES. Worrurncton G. Surra.—It is with much regret that we record the death of Worthington G. Smith, which occurred at Dunstable on 27th October, 1917. ‘To readers of the Gardeners’ Chronicle, Mr. Smith was best known as a_ botanical artist, and the clear, forcible illustrations initialed with the familiar ee W G ces ap 32 a view to- obtaining a figure for his own series of drawings of British fungi. ». Worthington Smith was born in 1835, and was a contemporary of Berkeley and Cooke; by his death, therefore, mycologists’ are deprived of a most interesting link with the past. His principal published works were:—Diseases of Farm and Garden Crops (1884), a supplementary volume to Berkeley’s Outline of British Pung? (1891), Man, the Primeval Savage, and the well-known and useful Synopsis of British Basidiomycetes (1911). Amongst other awards received for his work as an artist was the Gold Pe of the Royal pcre os and the Silver Medal oi the R.H.S. of Ireland. In 1865 he was awarded the Kaighnn Medal of the former ma for his. researches on Potato Blight; the results of this work, however, have not been altogether accepted by later botanists. Worthington Smith was the first freeman of Dunstable, and was President of the British Mycological Society in 1903. He was awarded a Civil List Pension in 1902 for his services to archaeology and botanical illustration. Epwakp Joun Woopuouse, Lizrur.—We record with regret the death, from wounds received in action in France, of Lieut. EH. J. Woodhouse, M.A., F.L.S., Economic Botanist and ered age of the ioe ‘College, Sabou, Bihar and Orissa, which oc- curred at a ess ge Clearing Station in France, on December 18th, 1917, aged 33. In Ceichas 1907, he was appointed Economic Botanist to the in the Bikar Tight lowed to join the Indian A Army Reserve of Oficers and in eA3 ruary, 1915, was attached to a cavalry regiment on the N.W. Frontier. In the following July he went to France, where he was attached to another cavalry regiment and became signalling officer, acting adjutant, and ‘officiating & sieeon commander. 33 Hildebrand experienced some difficulty in procuring seeds ot this plant owing to the fact that its fruits, which are nearly as large as a walnut and have a sweet pulp, are eaten by rats, squirrels and birds. However, he at length succeeded in getting some seeds which were received at Kew in September, 1892, but these failed to germinate. From a consignment received in September, 1893, some plants were raised, and in January, 1894, a Wardian case containing 49 seedlings of the Lonicera reached Kew, from which some were distributed to various establishments, including the Royal Botanic Gardens, Glasnevin, where the plant first flowered in August, 1898. In 1899 a plant flowered in the Temperate House at Kew, and material obtained from this was used in the preparation of the plate in the Botanical Magazine (t. 7677). Lonicera Hildebrandiana was afterwards discovered at Mengtze and Szemao in Yunnan, by Mr. Augustine Henry, but in the Chinese specimens the corolla is rather shorter than in the Burmese. It is only in very favoured localities, such as Cornwall, that this honeysuckle will thrive out-of-doors in these islands, and only a greenhouse of large dimensions will accommo- date so robust a grower. It is, therefore, never likely to become a popular plant. This remark applies still more forcibly to Rosa gigantea, Collett (Bot. Mag. t. 7972), which, though intro- duced into this country some years before Mr. Hildebrand com- menced his correspondence with Kew, was the subject of some interesting comments in his letters. We learn from these that in Burma the rose sometimes forms a bole as thick as a man’s thigh, and its stems overtop trees 60 to 80 ft. high, when, its branches hanging from the tree-tops, it blossoms profusely, its flowers appearing as a sheet of white and filling the air around with a delightful fragrance. Amongst other plants sent to Kew by Mr. Hildebrand were Bulbophyllum comosum, Collett & Hemsl. (Bot. Mag. t. 7283), originally introduced into this country by Colonel Sir H. Collett in 1889; Aeschynanthus Hildebrandii, Hemsl. (Bot. Mag. t. 7365), and Orchis monophylla, Collett & Hemsl. (Bot. Mag. t. 7601). Perhaps his best introduction was Dendrobium Hilde- brandii, Rolfe (Bot. Mag. t. 7453), which he sent to Messrs. Hugh ow & Co., of Clapton, who forwarded plants to Kew. This is a handsome and remarkably floriferous species allied to D. tor- tile, Lindl., with light yellow flowers, or, in the variety oculatum, with two maroon blotches on the disc of the lip. It is a parent of several interesting hybrids, including D. Elwesii, D. Ellisii an . Haywoodiae. to him from Kew. Mr. Hildebrand retired in March, 1902, and returned to his English home at Devizes, Wiltshire. He died at Puddletown, near Dorchester, at the age of 65. | | gies ¢ 34 oeeorag to Gardens.—The war continues to affect the ex- change, e between Kew and other botanical establishments. Of the Gonpucaeelt few consignments of material sent from Kew to oversea Ao ibe a few were "Tost in transit. This was also the fate of some consignments from abroad intended for Kew. Raebihistandiay these difficulties, many consign- ments of seeds and a few living plants were received from botanical and other institutions ali over the world. The prin- cipal receipts during the year were :— Botanic Gardens and other institutions: Loanda, Angola—Seeds of Palms and various plants of economic interes Arnold Arboretum—Many packets of seeds. South African Department of Agriculture, Pretoria— Various seeds, tubers and plants, including Teff grass (Eragrostis abyssinica ca). t of Horticulture, Cairo—Various seeds, including Hyoscyamct muticus, ‘Tle« paraguayensis and Hyphaene thebaic Dunedin Botanic Garden — Seeds of Celmisias and other New Zealand pla Hong Kong Won. Garden—Seeds of Rhodoleia cham- pom, Tutcheria spectabilis, Strychnos angustiflora, and Momordica cochinchinensis. United States Department of Agriculture, Washington— Various trees and shrubs. Singapore Botanic Garden—Various seeds, including lyphaene indica, Styrax benzoin Greenwich Park (Mr. iA Hay)—Gentiana Farreri, Meco- nopsis decora, etc. Honolulu, Board of Agriculture and Forestry—Seeds of alms, etc. Calcutta Royal Botanic Garden—Various seeds. artoum, snaked = Agriculture and Forests—Seeds oscyamus m Bac hee College Seay 5 eg Dublin—A morphophallus Nationa! Botanic Garden, oe Cape Town—Seeds of South African plants. Trinidad Royal Botanic Garden—Cacti from bos Island. Director Agricultural and Commercial Servi Sai Cochin- China—Seeds of Salida n, "peli from Uganda chips ahaa Bore Department—Seeds__ of Carapa grandiflora, e Dominica Botanic isdn: thai of Yams and Aroids. Royal Botanic Garden, Edinburgh—New Primula from China. Kumaun Botanic Garden species of Anoectochilus sp. Receipts from other sources included the following : — Mr. F. Nutter Cox, San José, Costa Bica Seeds of an indi- genous cotton, believed to be a new species * 30 Mr. M. T. Dawe, Director of Agriculture, Colombia—Seeds of rare and new plants, including Befaria sp. and Cacta- Mr. H. J. Elwes, Colesborne—Various plants and seeds. Miss A. Balfour—Seeds of Eucalyptus whittingehamensis. Countess of Selborne—Seeds of Meconopsis from Sikkim. Messrs. rar. and Co.—Seeds of Cybistaz antisyphi- sn oe Boliv ae Williams, Caerhay s—Chinese Rhododendrons and Penvdles er. ey. Perez, Teneriffe—Seeds of Palms, Pines, Echiums and Cytisi for distribution Prof. A. Henry—Seeds of Picea sitchensis, Larix alaskensis and Cedrus brevifolia. Mr. N. E. Brown, Kew—Stapelias, Mesembryanthe- mums, etc. Mr. A. Jackson, H.B.M. Consul, Madrid—Seeds of five species of Junipers from Madrid and Valencia, for dis- tribution. Collected by Mr, Alfred Martinez, of the Forestry Board, Madrid, ae Mr. Rafael Janini, Agri- cultural Engineer, Valenci Hon. Vicary Gibbs, eprmenin trees and shru Capt. J. Bush, Chatham—Pelargonium Cotyledonis fice St. Helen Mr. R. ValieiGih: Falmouth—Ferns from the Falkland Islands. M. Vilmorin, Paris—Hybrid Columnea. Mr. C. U. Stuart, Alaska—Seeds of native ate etc. : Watti Mr. Pp. “ay Williams, St. Keverne—Primula Maclaren Mr. R. Farrer, Lancaster—Collection of Chinese Plants (52 species). Mr. W. A. Milner, Sheffield—Primula Reidit. oa = W. Damon, Argentine—Seeds of Ilex paraguayensts Ma oh, C. Butter, Stellenbosch—Seeds of Telfaria pedata. Miss P. Wake, Northampton—A large Platycerium. Eo a F. R. S. Balfour, Dawyck, N.B. —Various seeds. Mr. Faleons. Prof. O. “Beccari—Seeds of Copernicia australis. Sir E. G. Loder, Bt., Leonardslee—Various trees and shrubs. Capt. G. H. Johnston, Trewithan—Seeds and bulbs from ‘tae _W. E. Loder, Ardingly—Various trees and shrubs. cz 36 Mr. C. D. Graham, Kildare—Seeds from Macedonia. Mr. G. Elisha, Canonbury Park—Mesembryanthemums. Mr. T. H. Burroughs, Stamford—Various trees and shrubs. Prof. J. H. Wilson, St. Andrews—Hybrid Passifloras and Begonias. Mr. A. Monk, New Jersey—Seeds of Gentiana crinita. Sir I’. Crisp, Bt., Henley—Various economic plants and palms. Major Howard, Richmond—Seeds from Kilamatende,. East ca. Fi commerce; Ilex paraguayensis,' which yield Paraguay tea; Clematis Stanleyi; Pouteria suavis; “Tefi’’, (Eragrostis abyss- inca); Befaria sp. from Sierra Nevada; Juniper seeds from Spain and seeds of Cytisi and Pinus canariensis from Teneriffe. . A number of plants raised for British Military Cemeteries in France, including 250 Eucalypti, Canadian Maples, and other trees and shrubs were sent to the Director, Graves Registration - and Enquiries, G.H.Q., France. During the year a number of nursery borders and some flower beds in the public grounds were devoted to the cultivation of vegetables, such as potatoes, turnips, parsnips, cauliflowers, cab- bages, carrots and bees Arboretum.—The difficulties arising from the shortage of labour, to which reference was mad plants which have been time, labour and mone The grass on many of the lawns, in normal times kept closely mown, was left to grow, thereby contribufing to the large stack of excellent hay which, in spite of the dryness of the early summer, was secured. Little damage has been done by storm during the past year as compared with 1916, but two fine trees were destroyed by lightning—one a large sweet chestnut near the North Gallery, the other a common oak near the Azalea Garden. The first was barked so thoroughly that it died by autumn, whilst the trunk got together with such an expenditure of . 37 of the latter was split to the very base. A spreading beech of huge dimensions near the Oxenhouse Gate was so badly wrecked by wind that it had to be removed. It was no doubt one of the comparatively few beeches that remain of the Rich- mond Gardens of Queen Caroline’s time. d The Chinese barberries introduced by Messrs. Wilson and Forrest are so numerous and have grown so well that it has been found necessary to enlarge the area devoted to the Natural Family to which they belong. Fresh ground has therefore been taken for them on the northern slope of the Flagstaff Mound. These new barberries are proving very ornamental in their fruits, and will certainly be valuable acquisitions among hardy William’s Temple has been re-laid out. It is proposed to plant here shrubs of exceptional interest or delicate growth. In many Natural Families there are species too small or not suffi- ciently robust to hold their own against the average run of their allies—such for instance as Pachystima Canbyi, Ilex Mariesii, Rhododendron Williamsianum and R. flavidum These, and others of a similar class, can here be given special care and kept free from the encroachment of greedy neighbours. At the request of the military authorities the Assistant Curator has visited parts of the coast to advise on the planting of sand dunes and for purposes of camoufiage; also as regards planting for shelter and ornament near new residential quarters being erected by the Admiralty. Very few additions of living trees and shrubs have been made, but acknowledgments are due to several amateurs who have contributed new species, especially to Mr. J. C. Williams, the Hon. Vicary Gibbs, and Mr. eginald Farrer. One con- signment only has been received from overseas, viz., some new plants, mostly of Chinese origin, from the U.S. Dept. of Agri- culture. Aircraft Park, Bristol; British Military Cemeteries in France. The following new rhododendrons have flowered for the first time at Kew:—R. adenopodum, R. hippophaeoides, R. olet- folium, R. polifolium, R. prostratum, and R. Sargenta j i which first The remarkably late flowering f#. auriculatum, blossomed at Kew in 1916, did so again in July, 1917, and much more freely than before. W.J. B. cate material for distribution has been limited. A large number of miscellaneous products have been received for deter- mination from Government Departments, scientific institutions, 38 . Museum No. 1. The buildings that are closed, due to “isle. _ tion of the staff, have in all cases been made av ailable to visitors desiring to examine the collections for special purposes or for general study. J. M. Presentations to Museums.—The following miscellaneous specimens have been Horde in addition to those previously recorded in the Bullet Crown Agents for ae Colonies.—Sample of Cocoa from the ameroons. Director, Botanic Gardens, Singapore.—Sections of wood of spe set al Mya tag Cupania pallidula and Elaeo- carpus Hullet Dr. Perez, Be ital ‘Ursula, Sepa —Roots and pieces of trunks of Convolvulus scopariu Mr. M. T. Dawe, Director of Agriculture, Colombia.— Specimens of native wood carving, seeds of Ivory Palm (Phytelephas macrocarpa), bark and resin of Cercediwm Mpg ha pods of Divi Divi (Caesalpinia coriaria), ete. . J. 8. Gamble, F.R.S., East Liss, Hants.—Seven small eudplse of varie les ny. mre Bo Ws ‘ mberton, Cesta Reserve Battalion.— Twelve ph piedioipenphs of root and branch grafting in British Mr. G. M. F.L.S.—Series - photographs of the Mahwa tree ae Apa fe gn H.M. Consul, Costa Rica.—Frui ts of The Balsa tree (Och- opsus Mr. Cecil Psi Kew.—Gnarl on Cedrus Libani Sir E. G. Loder, Bart., Horsham, Sussex. —Cluster of a of Pinus. T. Drewry, est Gate, Essex.—Section of a pile of Ballah wood (Shorea collina ?) bored by teredo, drawn up in Keppel Harbour, Singapore, 1907. Mr. E. R.. Pratt; Dewahar: Norfolk. —Planks of Abies Nordmanniana, Cupressus Lawsoniana, and a transverse section of Picea eacelsa. bot At. poe et Assistant Director of Horticulture, ‘cai gaia of Doum Palm (Hyphaene thebaica), c Mr. ee allace, Boscombe, Hants sone samples of Curare Scan from the Amazon Valley 39 Mr. C. U. Stuart, Seattle, U.S. America.—Fruits of Mehul pauciflorum from Alaska. A Waterer 2 PEP Hill Nursery, Woking.—Fruits of Og malus Colur The Hon. Visedy" Gibbs, Elstree, tured from Populus deltoides. a F.-D. Hartord, €.V.0., Henbury, Bristol.—Photo- aph of a tree of Ginkgo biloba growing in the grounds ee “Blaise Castle. Height 70 ft., girth at 4ft. 6 ins. from the ground 10 ft. 3 ins. Mr. H. C. Baker, Almondsbury, Gloucester.—Fruits ot Akebia quinata and of A. lobata. Messrs. F. W. & E. Bloore, Timber Merchants, Fern- dale er ad, Clapham.—A finely figured slab of Mahogan Messrs. Wi gglesworth & Co. Fenchurch Street London. Eee ls of Flax epeha usitatissimum), from British East Africa. Mr. H. Kempshall, Abbotsbury Gardens, Dorchester.— i bat a. Mr. C. Warren, Curator, Agricultural Department, Nigeria.—Abnormal fruit of Cocos nucifera and a cake of Camwood prepared from Baphia nitida, at Banoara, Southern Nigeria. 5. M. H. Research in Jodrell Ccbiantauy in 1917. —Mr. J. Bintner seo at: his observations on the development of callus-tissue and the formation of roots in cuttings, and began a study of some » gale on the stem of one of the Proteaceae A. Boodle studied some of the properties of charred ee and was engaged in anatomical w ork of various kinds. Mr. A. Sharples began an investigation of Hevea-wood attacked By. Ustulina zonata Mr. W. Worsdell examined the comparative anatomy of Compositan, Polygonaceac, etc. Pathological Laboratory. sig peek “feature of the season 1917 was the marked increase of interest in fungus and insect diseases of plants, and the recognition by a large body of the public of the amount of damage which various plants pests are capable of inflicting. some of the southern and south-wést districts bad attacks by way Oh on fruit trees in the spring, severe scabbing of pietnses ne losses, and large growers an market well as amateurs appealed for advice and for ane ceiheds of soniel, both by letter and by visits to the laboratory. With ett to the staff, Mr. W. B. Brierley was absent on military duty from _ January to the end of March. "ay L. Alcoc ms ~ nee ted permanent assistant, and has been engaged p lay on diseases of fruit trees. Miss M. N. Owen was pata temporary technical assistant in January. 40 I.—Apvisory Work. Some 800 enquiries were received through the post, which represents an increase of nearly 300 over 1916. 99 different kinds of disease were represented, including 14 diseases of pota- toes, 12 of apples, and 10 of cereals. Since the middle of August numerous questions relating to diseases received by the ings similar to blight and has been often confused with it. The fungus causes very little damage, and it was recommended to plant the affected tubers even though they had in some cases been previously condemned. he Black Leg of Potato, Potato Scab on newly-broken ground, Plum Rust, Die Back of Apples, Onion Diseases (Sclerotinia and Botrytis) and Sooty Blotch of Apples. Special mention should, however be made of. the following : — Black Currant Rust. (Cronartiwm ribicola).—Attention was called to this disease early in the year owing to legislative action by the United States of America. At the time the Black Cur- tant Rust was supposed to be rare and local, and a memo- randum was sent to the Board of Agriculture to this effect. The rust, however, was discovered in several districts about the middle far north as Norfolk. In some cases it was extraordinarily abun- dant and caused severe defoliation. No outbreaks occurred at how a disease may suddenly become epidemic after having existed in a country for many years in scattered localities. The appearance over so wide an area in 1917 suggests, however, that owing to the lack of an adequate phytopathological service the fungus had spread unnoticed during previous seasons. Onion Disease.—In addition Mildew, reports were received from several quarters that the onion crop was being badly at ini This disease, though long known, has never been properly 41 investigated, neither had the species been correctly determined. The fungus has been isolated, and it is proposed to work out its hfe-history, whilst Mr. Brierley will investigate the Botrytis with which it is often accompanied. The whole question of the diseases due to Sclerotinia and Botrytis is urgently in need of thorough investigation. ; . Il.—Researcu. American Gooseberry Mildew. (Sphaerotheca mors-uvae).— in continuance of the work on the dehiscence of the perithecia and the vitality of the spores of Sphaerotheca mors-uvae, on which a report was drawn up last year by Mr. Cotton and Mrs, Alcock, a number of bushes were planted in the Experimental Ground, and much material was collected The mildew, how- ver, did not appear, and was completely absent in this neigh- bourhood in 1917. Parsnip Canker.—This research was undertaken in response to enquiries from the Evesham growers last January, who stated that hundreds of tons of roots were being lost in their district. The work was carried out largely at Kew, and an account of the disease will be found in the current number of the Bulletin (see p. 8). certain varieties is a pl of sheer good fortune which has d the country from a very grave situation. The matter should not, however, be left in this ition, as immunity results so far obtained have been conflicting and chiefly negative, and the work has had to be set aside. The problem is a large and important one, and a thorough investigation, along bio- chemical lines, is urgently needed, and should be made a whole- time study by a competent investigator. The small size of the Staff has hitherto prevented this from being undertaken at Kew, even were the locality suitable for the purpose. series of experiments, designed to show the amount of heat required to kill the resting spores was attempted in the yard attached to the laboratory, but owing to the controls becoming infected the results were vitiated. The yard arrears now to b thoroughly contaminated with wart disease spores. Apple Mildew (Podosphaera leucotricha).—The research on this disease was concluded. It was found that perithecia-forma- tion often took place in the older parts of the mycelium, but 42 in spite of this the ascospores are apparently of little importance in the life-history of the fungus, as it reproduces itself by means of hibernating mycelium. The disease, therefore, must be controlled by pruning, and not by spraying, though experiments conducted at Wisbech and elsewhere showed that the conidial stage of the fungus could be very materially checked by the use of a strong lime-sulphur wash. A paper on the subject will shortly be published. Botrytis.—A comparative investigation of the morphology and parasitism of the fungus Botrytis cinerea has been carried out by Mr. Brierley, particularly in Aesculus Pavia and Ribes alpinum. An account of the behaviour of the fungus in the former will be found in Kew Bull., 1917, p- 315. In Ribes alpinum the fungus was found to be confined to the dead, water- conducting elements of the stem, and is present throughout the aerial portion of the plant; developing along with the host and existing in a symbiotic condition with its tissues. The host is stunted and stimulated to the formation of galls which morpho- logically are adventitious roots. The results of an investigation of the morphological characters of Botrytis cinerea and their specific value will be published in the Bulletin in the course of the year . Progress has also been made during the year in the investiga- tion of Thielavia basicola and Trichophyton crateriforme. few =. Sues of the commonly grown market varieties of fruit, e ITI.—Traveziine. Official visits were paid to the Wisbech district in connection with the spraying of apples for Apple Mildew and other diseases, and to Evesham and Worcester with reference to parsnip canker, Silver Leaf in plums, and obscure diseases of the potato. The spread of Weymouth Pine Rust was examined at Oxford, and visits were paid to Ormskirk and other localities in connection with Wart Disease and other maladies of the potato. , Presentations to the Library during 1917.—Prof. ©. S. ‘Sergent, Director of the Arnold Arboretum, has continued to send to Kew the official publications of the Arboretum, and these 43 of 1917 include The Conifers and Taxads of Japan, by EH. H. Wilson, and the final part of The Plantae Wilsonianae. The last-named work is in nine parts, forming three thick octavo volumes, comprising altogether nearly 2000 pages. Amon Prof. Sargent’s other presentations is a copy of Color Standards and Color Nomenclature, by R. Ridgwa Prof. Hans Schinz has sent No. 76 of the Mitteilungen aus dem botanischen Museum der Universitét Ziirich rom the Conservator of Forests, Western Australia, have been received A Discussion of Australian Forestry, by D. E. Hutchins, and a third edition of Western Australian Timber Tests, 1906 (and supplement), by G. A. Julius; and from the Trustees of the British Museum, parts 1 and 2 of the Botany of the British Antarctic Expedition, 1910. These deal with the Algae. Prof. L. H. Bailey’s Standard Cyclopedia of Horticulture has been completed by the issue in June of the sixth volume, and the establishment is indebted to the author and the pub- lishers, Messrs. Macmillan, for an entire set of this extremely useful work. y Centi y y of sented to the library. Ina letter to the Director, Dr. Prosper p. 347). The volume includes portraits of Cavanilles and La Gasca and numerous biographical notices of contemporary botanists. Notwithstanding the extremely unsettled state of Russia there is evidence that some progress in botanical research has been ade, and numerous publications have been received from Mr. Boris Fedtschenko, Mr. D. Janischewsky, Mr. H. A. Naumoff, and the Director of the Botanic Garden of Nikita, Yalta, Crimea. Mr. Fedtschenko’s presentations include an illustrated Flora of Turkestan and the Kirghiz Steppes, and portions of a Flora of Asiatic Russia, both works being in Russian. lished by the Netherland Botanical Society, vol. i. s (Monocotyledons), has been received from the Secretary of the ociety, and the four parts issued of the Communications from the Netherland Government Institute for Advising the Rubber Trade established. at Delft, have been received f the 4 Council of the International Association for rubber cultivation in the Netherland Indies. Prof. E. D. Merrill, of the Section of Botany, Bureau of Science, Manila, has been engaged for several years on the study of Rumphius’s Herbarium Amboinense, which had been begun by Dr. C. B. Robinson, whose lamented and tragic death occurred in Amboina on Dec. 5, 1918. Prof. Merrill has now published the result of his researches in a volume of 595 pages, entitled: An interpretation of Rumphius’s Herbarium Amboin- ense, a copy of which he has presented to the library. In the course of his investigations, Prof. Merrill has discovered over a hundred names, a list of which appears in his work, which have been omitted from the Index Kewensis. Dr. N. L. Britton has sent the parts published during the year cf the North American Flora, and the Director of the Depart- The following have also been received: Travauae du Labora- _.toire de Matiére Médicale de I’ Ecole Supérieure de Pharmacie de Paris, published under the direction of Prof. E. Perrot, vols. v.-x.; A Contribution to the Phytogeography and Flora of : N.W. N uinea), by L. S. Gibbs, and The Australian Flora in Applied Art, part 1, The Waratah, by R. T. Baker, from the Director, Royal Botanic Gardens, Kew; A Record of Plants collected in Southern ductive Industries, Government of Formosa; Gleanings in old garden literature, by W. C. Hazlitt, from Miss M. Smith ; Vernacular list of Pecos Shrubs and Woody Climbers in the Madras Presidency, by A. W. Lushington, from Mr. J. § The more important of the publications added to the library e:—British Grasses and their griculture, by S. F Armstrong, (from the 45 second edition; Monographie du Genre Melampeecn, Bi, by G. Beauverd; Marine e Algae of the Danish West Indies, by F. ded, vol. iii. pp. 145- 240; La Foret et les Bois du dias; by H. H. Haines ok the author and ne pestle e the Commissioner of the Central Provinces); Catalogue des Phanérogames de UArgentine, by IL. Hauman and Vanderveken, and 3 pamphlets by L. Hauman; The Trees, Shrubs, Herbs and Climbers of Sierra Leone, by C. E. Lane- Poole (2 copies) ; ih Poliporaceos de la Flora Espatola, by B. Lazaro é Ibiza; Bitter Pit investigation, fifth report, by D. McAlpine tentn the author and. Director of Quaratihe: Melbourne); A critical apes of the Genus Eucalyptus, by J. Maiden (pati 28 and 31;A monograph of the Genus Bric- hellia, by B. “e sbetienter: The Ohia Lehua Trees of Hawai, by J. F. Rock < copies) ; Hand book of the Bryophyta of South —— by T. Sim, parts 1 and 2; Seeding and Planting, by ciWss “hs nate and the Sieteenth report of the Woburn aie ae Fruit Farm, by the Duke of Bedford and Spencer Pickerin To the collection of travels has been added E. S. Meany’s work on Mount Rainier, which has been presented by Mr. Hervey Lindley, and amongst the peeotiedl publications re- ceived are several parts of the Annals of the South African Museum, from the ‘Trustees, and the gaa oe V Institut- Oinoue, Tokio, extra number, from Mr. Y. Manuscripts received include a valuable ition of 49 letters from Dr. F. Welwitsch, the celebrated explorer of Angola, to ‘the late Prof. Daniel wees written, judging from those bearing dates, between Sept. 2 1864, and Dee. 7, 1868, part of the period ‘duri ng which Welwitsch was resident in London. These letters have been presented by Prof. F. W. Oliver. A typed copy of A List of Plants found in the District of Uitenhage... together with a description of some new pene, y C.F. Feklen n, Additions to the Herbarium during 1917.—During the year about 6000 specimens were received as donations or exchanges, 1559 were purchased and 888 were received on loan. The prin- cipal collections are enumerated Evrore.—Purchased: Dr. J. Ww. Ellis, Drawings of British Fungi; H. Su ris Herbarium Hierasorum, fasc. Nort ca.—Presented: Egypt (F. G. Welanehiny aM the Ministry of Agriculture, Cairo; Sollum, by Capt Williamso CHINA. 5 einen tid China (G. Forrest and F. K. Ward), by the Royal Botanic Garden, ior ee gh; China and Japan (E. H. Wilson), by Prof. C. S. Sar, 46 Inpia.—Presented: Madras, by the . Madras Government Herbarium through Mr. J. 8. Gamble; Malay Peninsula, by Mr. I. urkill, and the Federated Malay States Government through Mr. C. Boden Kloss Mataya.—Presented: Siam, by Phra Vanpruk, Khun Winit Wanadorn, and the Federated Malay States Government through Mr. C. Boden Kloss. AusTrRALiA.—Presented: Kucalypti and Acacias, by the Director of the Botanic Gardens, Sydney; Mosses by Mr. W. Greenwood ; Bellenden Ker Range, by Miss L. S. Gibbs. Trorican Arrica.—Presented: Uganda, by Mr. J. D. nga, by Mr. A. Ringoet; Southern Rhodesia (J. A. S. Walters), by the Department of Agriculture, Rhodesia, and Mr. F. Eyles; Rhodesia, (A. J. Teague) by the Bolus Herbarium. Purchased: Rev. F. A. Rogers, Rhodesia and the Belgian Congo; R. A. Dummer, Uganda. Sourn Arrica.—Presented: Mosses and lichens, by the South African Museum. Norta Amertca.—Presented: British Columbia, by Miss E. en; United States (H. Eggert and others), by the Missouri Botanical Garden. CenrraL AmeERica.—Presented: Mexico (Fréres G. Arsene and Nicolas), by Prince R. Bonaparte. South AmeErica.—Presented : Colombia, by Mr. M. T. Dawe; British Guiana, by Mr. A. Leechman; Peru (Sr. N. E. Esposto), by Prof. Jules Gaudron. Plants have been received from Mr. W. B.-Turrill, Assistant in the Herbarium, who has collected them while on active service. The Chinese plants collected by Messrs. Forrest and Ward con- tain interesting specimens of Rhododendrons. The specimens received from the Federated Malay States Government contained plants collected in the Langkawi and Telebin Islands by Mr. C. Boden Kloss. Mr. J. M. Black has presented specimens of new species of Australian plants described by him in the Transactions of the Royal Society of South Australia. Miss L. S. Gibbs has given specimens collected by her in the Bellenden Ker Range. Some of Mr. J. D. Snowden’s plants came from Mt. Elgon. An interesting collection from Katanga containing many plants of economic interest has been received from Mr. A. Ringoet. Miss . M. Warren has sent additional specimens from the Kootenay region of British Columbia. - M. T. Dawe’s Colombian col- lections have been made along the Rio Magdalena, Santa Marta, Cartagena and the Goajira Peninsula. Prof. W. Bateson has presented South American Calceolarias and hybrids produced from o important sets of drawings have been received : one of British Fungi from the collection of the late Dr. J. W. Ellis ( K.B. 1917, 87); the other of Orchids drawn by Mrs. Janet Ross and presented by the Bentham Trustees (See K.B. 1917, 85). 47 Hippocratea Cumingii and H. Maingayi._An enquiry as to the original specimen of Hippocratea Cumingii reveals an amount of confusion that can bid be dealt with in a short note. Hippo- cratea Cumingu, Laws., was described in 1875, being based upon flowering specimens collected i Pe Malacca by Griffith, and in the Philippines by Cuming (no. 1725), the ‘specific name being derived from the latter. Sir George ruck afterwards enumerated it as an imperfectly known species. It proves on comparison to be identical with the earlier H. heiptig ha Korth., based upon a specimen collected at Banjermassing, Borneo. The latter name was untortunately transferred by Sir George King to another plant, with the remark, ‘‘ flowers of Chittagong and Assam speci- mens of this which I have dissected agree perfectly with Korthals’ figure.”” This figure, however, shows villous petals, and the other characters are also not in agreement. He also cites as synonymous, H. lanceolata, Ham., H. “grandiflora, Wall., and H. obtusifolia, Laws. (in part, not of Roxburgh), with the addi- tional localities, Sikkim, Burma, South Andaman and Pera These, unfortunately, do not represent the plant of Korthals, and, further, the Perak plant is distinct from the others cited, an nd been described under the name of H. nigricaulis, Ridl. It also includes a specimen from Malacca, Griffith, which is written up as H. obtustfolia, Roxb., though not cited by Lawson, so that it is uncertain whether he intended to include it. There is also the difficulty that both Wallich and Lawson included under H. obtusi- folia, Roxb., materials which do not belong to it, as pointed out by Sir George King. This point, however, has not yet been fully cleared up. The Perak specimen just mentioned, however, belongs to ok laiamiat Ridl. The synonomy and distribution are as follow H. scala Korth. in Temminck, Verh. Nat. Gesch. (1839- 42), z. ane t. 7 ie King). H. Cumin ingit, Laws. in Hook. f. Fl. Ind. i. p. 624 (1875).—Ceylon, Walker; Malay Penin- rier "Malaves ‘Griffith; Kew Distrib. 906; Johore, Ridley 4183, 11139, 13446; Borneo: Banjermassing, Korthals; Kuching, Haviland 2859 ; Kulong, Haviland 2876; Sarawak, Beccari 4063 ; Philippines : Island of Samar, Cuming 1725. H. nigricaulis, Ridl. ined., H. macrantha, King, Mat. FI. Malay Penins. i. (1896) p 643, in part, (excl. syn.; non Korth.). —Malay Pensa: Matus, Griffith; Perak, Kite Collector 5118, 7570, Wray 631; Siostechiné 1047. The two plants are readily distinguished. H. macrantha has Stacy acuminate leaves, and fetals densely glandular- villous on the inner surface, while H. nigricaulis has obtuse, or very er and abruptly acuminate leaves, and the petals are labro gs Hippocratea Maingayi, Laws., was based on a fruiting speci- men collected at Malacca pe Ae egy dee see the eed “* ? Borneo’’, the source of w o. 3005. Both, however belong to Lapoeaben sinc eam piney ‘which 48 is based on a flowering specimen from Malacca, Maingay. The synonymy and distribution are as follows :— Lophopetalum reflexum, Laws., in rs Brit. Ind. i. p. 616 (1875), King, Mat. Fl. Malay Penins. 1. p. Retin Hippocratea Maingayi, Laws. in Hook. f. Fl. Brit. Ind p- 625 (1878).— Malay Peninsula, Maingay, Kew Distrib. 3933/2, 397; Penang, Curtis 1502, 3607; Borneo: Sarawak, Beccari 3005. ye tee Strychnos angustiflora.—The button-like seeds of this Hong- kong Strychnos are so similar in their general appearance to those . Nux-vomica that it seemed desirable to have them analysed. Mr. Tutcher, Superintendent, Botanical and Forestry Depart- ment, Hongkong, to whom, we are indebted for a supply of seeds, also informed us that he had recently received an application from a Chinaman to cut down all S. angustiflora plants growing near his house as the fruits dropped into a stream from w rhich he obtained his water supply and poisoned the water. Prot. Greenish, who has Kndty examined the seeds at the laboratories of the Pharmaceutical Society informs us that work- ing by the Pharmacopeeia method the residue which would con- tain all the strychnine and brucine amounted to only 0028 per cent, and gave only a slight reaction for alkaloid. No trace of strychnine or brucine is present in seeds. The poisonous property alleged to be present in the plants grow- ing in Hongkong may therefore reside in the fruit or the asser- tion may peers to be unfounded. [Crown Copyright Reserved. ROYAL BOTANIC GARDENS, KEW. BULLETIN ~ MISCELLANEOUS INFORMATION. Nos. 2 & 3.] [1918 VII.—_THE GENUS CHROZOPHORA. D. Pra. The Euphorbiaceous genus Chrozophora is so easily distinguish- able from its nearest allies as to suggest that it is a ‘natural ’ one. The divergence of view that has prevailed as to the limita- tion of its component forms has been so striking as to indicate that it is a ‘di It’ one. These forms, Bentham remark in 1880 (Gen. Plant. iii. p. 305) are ‘ inter se valde confusae non- nullae revidendae et forte iterum dividendae vel reducendae.’ In a serious attempt to supply the revision thus advocated, which was published by Pax and K. Hoffmann in October, 1912 (Pflanzenr. IV. 147 vi. pp. 17-27), the species recognised have been said to be ‘ inter se saepe simillimae et caute examinandae.’ This was the experience met with when an account of the African species, also, as it happens, published in October, 1912 (Fl. Trop. Afr. vi. 1, pp. 834-9), was being prepared. hile that study was in progress, it was found impossible to arrive at definite conclusions regarding some of the African forms until careful investigation had been made of the material available from Europe and Asia. This involved, in fact, the examination of all the specimens in a number of different herbaria, including those of Kew, the British Museum, the Linnean Society of London, Paris (including the Lamarck and Jussieu collections and that of M. Drake del Castillo), Brussels, Leiden, Copenhagen, Geneva, Berlin (including the Willdenow herbarium), and as regards Indian material that of Calcutta and that of Mr. J. R. Drummond. In this way it has been possible to see the actual marck, Forskal, Vahl, Willdenow, Roxburgh. : jeu, Bunge, Wallich, Baillon, Anderson, Dalzell, Schweinfurth, Miiller, Boissier, Hooker, Broun, and Cooke have ased. The only types not seen have been two each of Burmann and Visiani, and one each of Geiseler, Delile, and Presl; the present whereabouts of these seven types it has not been possible to trace. he recent monograph in the ‘ Pflanzenreich ’ renders unneces- sary the preparation of another. At the same time a survey of the material known to be available for study may be of use as (5214.) Wet. 152—699. 1,125. 4/18. J.T. &S8., Ltd. G.14. Sch. 12. 50 a supplement to the ‘ Pflanzenreich ’’ monograph, and may serve as a guide to the whereabouts of the material which it will be incumbent upon the future monographer of the genus to consult. History OF THE GENUS. The name Chrozophora was applied by Necker in 1790 (Liem. ii. p. 337) to a monotypic genus based on a plant from Languedoc which at that date was the source of one of the Litmus dyes known as Tournesol. This name has, indeed! been used by most botanists since the XVIth century for the plant itself. That plant had been referred by Royen in 1740 (I. Lugd. p. 532) to the genus Croton as defined by Linnaeus in 1737. In 1748 Linnaeus accepted this determination (Hort. Upsal. p. 290); it was adopted by his friend and correspondent Sauvages in 1750 (Monspel. p. 305) before Linnaeus enumerated it, with a definite specific epithet, as Croton tinctorium in 1753 (Sp. Pl. p. 1004). oyen was not the first writer to refer the Tournesol to a genus from which it is better kept apart. The plant was well known to systematic writers in the XVIth century. ome of these, convinced that words connoting the same idea should denote the same thing, connected the Tournesol now with one, now with another of the plants which the ancients termed Heliotropion. Of these the Greeks, according to Dioscorides, knew two sorts— ndLoTpOTLoY TO péya, Or SKopriovpov; and mAvoTpoTLov TO juKpoV. The Latins, according to Pliny, also knew two—Heloscopium, or Verrucaria; and Tricoccon, or Scorpiurum. In 1554 Castell-Branco, better known as Amatus Lusitanus (Diosc. Enarrat. p. 437), identified the Tournesol of Spain with WAoTpoTov To péeya, and repeated this identification with some insistence, by including also the Tournesol of France, in the larger edition of this work (p. 741) in 1558. This emphasis may have been due to the fact that in 1554 both Dodoens (Cruyde- boek p. xii.) and Mattioli (Comm. Diosc. p. 561), had suggested a different identification. It may have been intensified because Clusius in 1557, in his French version of Dodoens (Hist. Pt, p. ult.), had figured the Tournesol as Heliotropium parvum, and , published in 1561. Here Cordus has remarked of 7d uéya ‘ qualis sit herba et quomodo jam vocatur, ignoro cum omnibus ’; of To wuxpov: ‘quae sit hodie omnino ignoramus.’ This did not deter Gesner from identifying, in a treatise appended to that 51 moment, since neither of them is the Tournesol. What, how- ever, does concern us is that Gesner was satisfied that one of Pliny’s plants, ‘ Heliotropium minus quod et Tricoccum cog- nominant’ must be distinct from either of those described by Dioscorides. With ‘ Tricoccon’ Gesner identified the plant in the Padua garden which, we learn from him, had been raised from seed sent from Crete. This plant, which Gesner, from hear- say, believed to be wild near Montpellier, he considered unworthy of cultivation. Nevertheless, as we learn from Schmiedel (Gesn. Op. Bot. i. t. 4, fig. 30), Gesner made a drawing of the Padua garden plant, which shows that this plant was a form differing only in the outline of its leaves from the Tournesol of Spain and Provence, figured by Clusius. figured as H. minus tricoccum the plant known ‘castellanis Tornesol narbonensibus Tornesola quae voces,’ he added, ‘a graeco nomine deductae videntur.’ In the same year L’Obel, writing alone, reverted to the view of Pinaeus, and published the same figure (Hist. Stirp. p. 133) as ‘ Heliotropium parvum, Diosc.; H. minus Clusii; Triccon [sic] Plin.; Advers. p. 101.’ The erroneous typography was corrected in 1581 to ‘ Tricoccon ’ (Ie. Stirp. p. 261). When, in 1586, Camerarius issued a revision of Mattioli’s Com- pendium of 1571 he replaced Maittioli’s figure of H. minus by Gesner’s portrait of the Cretan plant in the Padua garden (Epit. Matt. Diose. p. 1001), under the impression that it represented the tinctorial one of Narbonne. Two years later Camerarius had ascertained that the Languedoc plant also oceurred in Italy— near Tresolza and elsewhere in Emilia, at Ortona and elsewhere in the Abruzzo (Hort. Med. p. 73). In the 1611 edition of his text. (p. 487a), grows in Syria as y The pee Mie ae which is that of the plant raised at Padua from Cretan seed, represents with accuracy a form of the Tour- nesol which, as Camerarius has stated, ‘is plentiful about Aleppo Tournesol to H. tricocewm, and if Bauhin in 1596 (Phytopin. 52 immediately after Mattioli’s figure of H. minus. In 1623 Bauhin again used the name H. tricocewm for the Tournesol century, was cited by Linnaeus in 1753 as a recognised synonym of his Croton tinctorium. Necker was not the first botanist to regard the Tournesol as the type of a distinct genus. A quarter of a century earlier Adanson had done this, as O. Kuntze has remarked (Rev. Gen. Pl. ii. p. 621), under the old French name Tournesol used by Magnol and latinised by Scopoli into Tournesolia, ‘welcher name’ Kuntze has added, ‘ auch von Necker citirt und nur wilkirlich in Chrozophora veraindert wurde.’ This, although a perfectly accurate statement, does not tell the whole story; agnol’s use of the name Tournesol was not published until after his death. The first author to accord generic rank to the Tournesol was Ray, who in 1686 dealt with the Heliotropium tricoccum ot Bauhin (Hist. Pl. i. p. 165) under a natural group of plants 4 tropium, he remarked (Elem. i. p. 116) ‘la plante qu’on apelle Heliotropium tricoccum n’est pas de ce charactére et doit faire un genre différent.’ s.’ This new suggestion was in turn taken up b Tourne- fort when, in 1700 (Inst. ret herb. ed. alt. p- 651), ie aetublished . his tinctorial species -paratur Tournesol ; tb i—‘nec potest dici Ricino affinis’—thus cancelling his own view of 1697, ‘nec Ricinoides ’—thus 53 In the meantime, however, new species were being added to Ricinoides, Tournef. In 1717, Hermann (Zeyl. 202, t. 11) reported one from Ceylon; in 1728, Martyn (Hist. Pl. r. cent. 1, ii. 38, 46) described two more from America. These, with others, were included, along with the Tournesol, in the genu Croton base y Linnaeus on the ruins of Tournefort’s Ricinoides, so that from the outset the Linnean complex wa even more confusing than that which it replaced. Notwith- standing all this, Moench, in 1794 (Meth. Pl. p. 286) proposed the revival of Tournefort’s name Ricinoides, as a substitute for the name Tournesolia of Scopoli. This attempt failed as those of Scopoli and Necker and Adanson had. The utility and convenience of his ‘ Species Plantarum ’ were so great and so undeniable as to make the authority of Linnaeus paramount, and the particular arrange- ment adopted in the case of Croton was followed, in spite of its imperfection, to mention only those authors who have added to our knowledge of Chrozophora, by Burmann (F/. Ind.) in 1768, by Forskal and his editor Zoega (FJ. Aegypt.-arab.) in 1775, b Lamarck (Encyc. Meth.) in 1786, by Vahl (Symb. Bot.) in 1790, by Willdenow (Sp. Pl.) in 1805, by Geiseler (Crot. Monogr.) in 1807, by Delile (FJ. Aegypt.) in 1812, by Sibthorp and Smith (F1. Graec.) in 1813, by Roxburgh (Hort. Beng.) in 1814, and by La Gasca (Gen. et Sp. Nov.) in 1816. So far as India is concerned, the Linnean usage lasted rather longer; raham Bomb.) in is however, was largely accidental. Graham’s local list was based o xburgh’s recently published work; Ainslie’s account was adapted from the names used by Roxburgh and his contemporaries and corres- pondents Buchanan (afterwards Hamilton), Heyne, Koenig, Rottler and Russell. When the generic status of Chrozophora, Neck., was re- vindicated by Ad. Jussieu in 1824 (Tent. Gen. Euph.), the spelling was changed to Crozophora. Jussieu’s action was accepted and endorsed by Sprengel (Syst. Veg. iii.) in 1826 ink, who recognised the apt in 1831 (Handb. il. p. 438), assuming, though the plant bears no particular resemblance to a 54 tassel or plume, that the etymology must be xpocaos and ¢gopos, wrote the name Crossophora. With this exception, the ortho- graphy of Jussieu and Sprengel was copied by Spach (Hist. Veg. 11.) and Decaisne (Florul. Sin.) in 1834, the younger Nees (Gen. Pl. Germ.) in 1835, Endlicher (Gen. Pl.) in 1840, Reichenbach (Ic. Fi. Germ. et Helv.) in 1841, Ledebour (FI. Ross.) in 1849- 51, Bunge (Rel. Lehm.) in 1851, Payer (Organogén.) in 1857,” Baillon (Etud. gén. Euphorb.) in 1858, Dalzell (Fl. Bomb.) and Klotzsch (Mossamb. Bot.) in 1861, Schweinfurth (Pl. Nilot. and other treatises) in 1862 and subsequent years, Miiller (DC. Prodr.) in 1866, and Boissier (Fl. Orient.) in 1879. _ A partial exception to the adoption of Sprengel’s treatment oecurred in 1836, when Visiani (Pl. quaed. Aegypt. ac Nub.) figured two species of our genus, one as a Chrozophora, the other as a Croton. Schweinfurth has suggested (PI. Nilot. p- 12) that this segregation may have been due to imperfect characterisation of Croton by Necker and Endlicher. But Endlicher did not describe Croton till 1840, and there is nothing to suggest that Visiani went behind Ad. Jussieu. We know, however, that the stamens in the species which Visiani referred to Croton are 3-ver- ticillate, in that referred by him to Chrozophora the anthers are e-verticillate, so that Visiani may have been influenced by this character as much as by the calycine one to which he and Schweinfurth have alluded. Bee himself; it is a Caperonia. Presl’s further note reads :— 56 generic value; there are five distinguishable forms in the genus Chrozophora which share it. The reduction of Lepidocroton to Chrozophora eftected by Bentham in 1880 (Gen. Pl. iii. p. 305) seems, therefore, to be justified; it is at least free from the objec- tion which attends that proposed by Baillon in 1858 and accepted by Miller in 1866 (DC. Prodr. xv. 2, p. 751). When accepting Bentham’s reduction in 1912, Pax and Hoftmann (Pflanzenr. 1V. 147, vi. p. 17) regarded as the type of Lepidocroton another Kordofan plant, gnlleoted by Kotschy at Abu Gerad and issued by Hochstetter as Kotschy n. 25, C. senegalensis, under which species the authors of the Pflanzenreich monograph account both for this specimen and for Presl’s genus. ‘This conclusion is open to the objection that Kotschy n. 25 has only seven to nine anthers whereas Lepidocroton should have fifteen. We know, however, that Kotschy did collect at Wolet Medine in Sennar another Chrozophora which was issued by Hochstetter as Kotschy n. 473, in which the male flowers have fifteen stamens. The species in question, C. plicata, A. Juss., is the only African Chrozophora which exhibits the character relied on by Pres] in distinguishing his genus Lepidocroton from Caperonia on the one hand and from Chrozophora on the other. For this reason it is to C. plicata that, in an account of the African species of Chrozophora pub- lished in 1912 (Fl. Trop. Afr. vi. 2, p. 885), Lepidocroton serra- tus has been referred. P.Br. (1756) to that of Tournesolia, Scop. (1777). But neither Bentham in 1880 (Gen. Pl. iii. p. 305), nor Pax in 1890 (Pflan- there is much to be sa When in 1880 Bentham reverted to the orthography originally employed by.Necker he cona@luded that the etymology of the word all but two botanists, since 1826, its validity and its ortho- graphy are not likely to be further impugned. 56 PosITION OF THE GENUS. The position of Chrozophora within the Euphorbiaceae has In 1866 Miiller tions in one Euphorbiaceous genus. Bentham in 1880 (Gen. Pl i. p. 248) and Pax in 1890 (Pflanzenfam. iii. 5, p. 42) were unable to accept so drastic a change of The former was, hame in a sense more comprehensive than that of Miller. To Miller the Crotoneae constituted a group equal in status with the Acalypheae; to Bentham the Crotoneae appeared a group of higher status within which the Chrozophoreae, the Crotoneae proper (Lucrotoneae), and the Acalypheae enjoy equal rank. The most important difference between the view of Bentham and of Miiller therefore lies in the fact that the former has treated the Chrozophoreae and the Acalypheae as of equal importance. Pax has adopted a compromise between the two views. He has, like Bentham, merged the Caperonieae of Miiller in his Chrozo- phorinae, but has at the same time, like Miller, subordinated Chrozophoreae a status equal to that of the Acalypheae. So far as mere convenience is concerned it is, however, clear that there is nothing to choose between the systems of Bentham and of Pax, and that either system is more satisfactory than that of Miiller. CHARACTERS OF THE GENUS. So far only two endeavours have been made to analyse the genus. To admit of an appreciation of these attempts a sum- mary of the characters is necessary. Chrozophora, Neck. Flowers monoecious, dichlamydeous. $ Calyz closed in bud, ovoid or globose, ultimately splitting into 5 valvate lobes. Petals 5, usually dirty yellow,. rarely red, generally shorter than the calyx lobes externally lepi- dote. Stamens 5-15; filaments connate, at least below, in a cen column, the free portion usually shorter, rarely considerably longer than the anthers; anthers usually ot 2-seriate, 10 or fewer, less often 3-seriate, 15; cells parallel, contiguous; dehiscence longitudinal, extrorse. udimen- tary ovary 0. 2 Calyz-segments valvate, narrower than in he male. Petals narrow, usually much smaller than in the male, sometimes setaceous, occasionally obsolete. Disk composed of 5 short wide rather prominent glands alter- nating with the petals. Ovary 3-celled, clothed either with stellate hairs or with flat peltate denticulate or subentire scales; styles 3, erect or somewhat spreading, usually stout, always 2-fid; style-arms entire, usually red, rarely orange; ovules solitary in each cell. Capsule 3-dymous, at first some- what fleshy, usually tinctorial, when ripe red, violet purple, or white tinged with violet; occasionally non-tinctorial, when ripe grey or nearly black; pericarp smooth or tuberculate, clothed with stellate hairs or with flat pectinate or subentire scales, very rarely the scales obsolete. Seeds without a caruncle; testa smooth or somewhat rugose; albumen fleshy; cotyledons broad, flat.—Coarse herbs, usually monocarpic, less often wndershrubs and polycarpic, everywhere clothed with stellate tomentum or very rarely nearly glabrous. Leaves alternate, stalked, usually undulately toothed, plicate-rugose or plicate-bullate or nearly flat, often with two glands near the apex of the petiole beneath. Flowers in short, dense, sessile or stalked racemes in the upper axils, each solitary ct; the males higher up, rather close-set, short- pedicelled or subsessile, the females near the base, few, dis- tinctly or sometimes long pedicelled. The first attempt to break up the genus into groups was made by Miiller in 1866 (DC. Prodr, xv. 2, pp. 747-750). The primary sub-division was effected by separating the forms in which the anthers exceed 10 and are 3-seriately arranged from those in which the anthers are 10 or fewer and are 2-seriately arranged. All those forms in which the anthers are 3-seriate have stellate- pubescent but not lepidote capsules. Of those forms in which the anthers are 2-seriate, only one has stellate-pubescent but not lepidote capsules; this was accordingly separated by the character in question from the remaining forms, and thus provided Miiller with a second group. The forms with 2-seriate stamens and lepi- dote capsules were subdivided by him into two more groups, according to the nature of the seed-coat, those with rough seeds constituting Miiller’s third group, those with smooth seeds his fourth and last group. The arrangement thus was :— (1) Sect. 1. Stamina triverticillata, saepius 15. Ovarium pilis stellatis vestitum. Sect. 2. Stamina 1-2-verticillata, 5-10:— Sub-sect. a. Ovarium pilis stellatis vestitum. Sub-sect. (3. Ovarium peltato-squamigerum :— (3) ‘ Group 1. Semina tuberculato-asperata. (4) : Group 2. Semina laevia. Miiller did not suggest names for any of these groups. 58 The other attempt to arrange the components of the genus was made by Pax and Hoffmann in 1912 (Pflanzenr. IV. 147, vi. pp. 18, 19). Adopting the characters relied on by Miiller, these authors have treated the group with 3-seriate anthers and stellate pubescent ovaries as a distinct section, 1. Plicatae. Those forms with 2-seriate anthers and lepidote capsules were subdivided into two more sections. Pax and Hoffmann have treated Miiller’s last group, in which the seeds are smooth, as a distinct section, 2. Senegalenses; the forms with 2-seriate anthers, lepidote cap- sules and tuberculate seeds they have regarded as forming a third section, 3. Tinctoriae. As compared with the arrangement pro- posed by Miiller, that of Pax and Hoffmann supplies a section 1. Plicatae, corresponding with Miiller s section 1, with two se tions, 2. Senegalenses and 3. Tinctoriae corresponding, in reversed order, with Miiller’s two groups under Section 2, sub-section 3. quite consistently with the group which constitutes Miiller’s Section 2, Subsection a. In their careful Conspectus Sectionum (l.c. p. 19) these authors have been at pains to exclude the group in question both from their sections 2. Senegalenses and 3. Tinc- toriae, though they have elsewhere (l.c. pp. 18 and 21) included it under 3. Tinctoriae, with the re a members of which the separates it from their section 3. Tinctoriae, when they state that Miiller’s second group cannot satisfactorily be included in the Tinctoriae, and to throw doubt on the conclusion (1l.c. p. 18) that this group has been derived from their section 3. Tinctoriae: vertently tran generic description (l.c. p. 17) and in the conspectus (1.c. p. 18), th of the sections Senegalenses and 59 Tinctoriae (l.c. pp. 20, 21); based upon their own observations, repeated the fact that the anthers are 2-verticillate. There does not seem to be any exception to this; even in such extreme cases of reduction as result in the presence of only four anthers it is found that at least one of these belongs to a higher and more central whorl than the remaining three. The scope of the present paper is such as to preclude the con- sideration of the interesting speculations themselves. The varia- bility of the androecium on which they are based is, however, so great as to suggest that this portion of the flower is on that account less suitable as the source of a character on which to base a primary subdivision of the genus. It is clear, from the difficulty experienced by Pax and Hoffmann in dealing with the type of Miiller’s section 2, subsection a, that the new serial dispositions adopted by them is hardly an improvement upon that suggested by Miiller in 1866, and it is satisfactory to find that, if we employ for the purpose of primary subdivision that to which Miiller gave the second place, but which in the case of their section 3, Tinctoriae, Pax and Hoffmann have been com- pelled to disregard, we are able to maintain Miiller’s serial dis- position and to break up the genus into what appear to be fairly natural groups. Treating as a distinct section, 1. 7richocarpa, all those forms in which the carpels and ripe capsules are stellate-pubescent but not lepidote, and as another distinct section, 2. Ledipocarpa, all those in which the carpels and ripe capsules are lepidote but not stellate-pubescent, we are able further to subdivide Tricho- portion of their respective filaments; and (3. Graciles, the group treated by Miiller as his section 2, sub-section a, with 2-seriate Hoffmann, with purple. usually m e, capsules loosely clothed with pectinate scales, and with always tuberculate seeds; enegalenses, exactly as defined by Pax nn, subentire flat scales, and with smooth seeds ecog- nised as distinct species. These ‘species,’ it should be under- stood, are recorded ‘without prejudice’; the question of the have been distinguished by some competent authority, who has actually examined and compared representative specimens of closely allied plants. In cases where authors of equal authority have described a second time some form already adequately characterised, only the original name is given. An instance 60 of the first kind is supplied by C. sabulosa and C. gracilis, whee were carefully distinguished by Ledebour (Flor. Ross. iii. p. 581), but which we now believe to be conspecific. An example of the second kind is supplied by C. obliqua, which has been described as a distinct species, under four different names, by Willdenow, Sibthorp and Smith, La Gasca, and Bunge, since it was first characterised by Vahl. ConsPectus oF SEcTIONS AND ForMS HITHERTO PROPOSED. Carpels Sis aes aac but never lepi an —_ Trichocarpa. Pees sabe 3-verticillate longer than the free portion of the filaments ... {a. Plicatae. Capsules purple ; stigmas red :— acemes nearly to quite as long as the adjacent leaves :— Lanse ovate, hardly lobed ... 1. Roftleri, A. Juss. Leaves rounded, more or less - distinctly 3-lobed 2. Burmanm, Spr. sees no much shorter than RE ves : cartes wodulntes peduncles always very short, often leafy at the base.. 3. plicata, A. Juss. Leaves repand ; peduncles dis- tinct, always naked .. 4, obliquifolia, Baill. Capsules grey or blackish ; nee ea Leaves 2-glandular at the base... 5. parvifolia, Klotzsch. Leaves eglandular 6. prostrata, Dalz. Anthers 2-vetiilate, shorter than the free portion of the filaments; emit purple; = red EB, Graciles. Leaves all obtuse .. .. 7. sabulosa, Kar. & Kir. Leaves acute, rarely 0 use 8. gracilis, Fisch. & Mey. a lepidote but never stellate- nt; anthers 2-verticillate ... §2. Lepidocarpa. Onhieutes purple, their scales discrete, with pectinate a ; seeds tuberculate Vy. Tinctoriae. Capsules very distinétly iruticatis: _ Leaves e as long as broad; ripe sesemenee blue-purple ; > stem 9. oblongifolia, A. Juss. Leaves si ee ini iat ; Tipe carpels red- purple :— Stem erect :-— Leaves thinly pubescent ... 10. tinctoria, A. Juss. Leaves densely ened o» Ll. htero ierosolymitana, Spr. Stem prostrate .-. 12. subplicata, Pax & Hoftm. 61 Capsules slightly muricate or nearly sm : Leaves densely softly villous with ong hairs ... 40 ... 13. obliqua, A. Juss. Leaves almost glabrous ... ... 14, glabrata, Pax & Hoffm. Capsules nearly white, their scales contiguous or imbricately over- lapping, with subentire or entire margins ; oth ... § 6. Senegalenses. Leaves heterophyllous, dark-green and or nearly so above, closely hoary or laxly woolly beneath gfe ... 1d. senegalensis, A, Juss. Leaves uniform in outline, paler green, from puberulous to woolly above and from closel hoary to laxly woolly beneath.. 16. Brocchiana, Vis. History or THE Spgcres uNDER Rucrnores. When in 1700 Pitton de Tournefort placed the Tournesol in his genus Ricinoides, he termed it ‘ Ricinoides ex qua paratur Tournesol gallorum’ (Jnst. rei herb. ed alt. app. p. 655), a phrase perhaps more accurate than that of the authors of the one of the Cyclades. This second species Tournefort in 1703 published (Cor. p. 45) as Ricinoides ex qua paratur Tournesol gallorum, folio oblongo et villoso, taking the opportunity to amplify the name of the original species from Southern France to Ricinoides ex qua paratur Tournesol eye folio serrato non villoso. There is in the Jussieu herbarium an example of this plant collected in Melos by Gundel, which has been named by Tournefort himself. In the British (Natural History) Museum there is a specimen of each of these two species, both of them written up by Tournefort. et There is in the Jussieu herbarium a specimen originally in the herbarium of Danty d’Isnard, which has been written up by Isnard himself as Ricinoides memphiticus folio laevi, Lippi. As the leaves are pubescent it cannot be the species Lippi had in mind when he proposed that name, and a later note accompany- ing the specimen, written, it would, however, appear, before the specimen itself reached A. L. Jussieu, says:—‘Cette plante comparée avec la description de Lippi ne lui convient pas. The plant is an example of Chrozophora oblongifolia, A. J uss., a species which in Egypt is strictly confined to the Red Sea lit- toral. We know, therefore, that, even if there had been no in- superable objection on morphological grounds to the identifica- tion by Isnard, the specimen did not come from anywhere near Memphis. This, however, does not affect the possibility that it was obtained by Lippi; the fact that Isnard wrote it up as one of Lippi’s plants renders it probable that it had been collected, 62 or at least formed part of the herbarium accumulated by that unfortunate traveller. Lippi accompaniec enoir Duroule uring ae Nong at! of the latter to the Court of Abyssinia in 1704-5. r landing at Alexandria Lippi made various ex- cursions in bute Egypt before he set out on the Abyssinian journey, in the course of which he was murdered in 17 4. known that he was able to visit Rosetta, Cairo, and Assiout, and although we are not told that he visited Suez, we are aware that if he did make that particular excursion, ot if he sent a collector there, he could hardly have failed to find, in the neigh- beurhood of that town, the species attributed to ae by Isnard. Some twenty years later yet another species of the genus was collected in India by Garcin, then a surgeon in the service of the Netherlands East India Company. On 11th August, 1722, aa left Batavia in a bps boun tir Surat. Garcin ‘landed Saetih wrote :—‘ La Saivall ou je me * péncontiay dans Suratte n’estoit point gies des plantes parceque la Secherease qui An dea getirg roit brulé toute la Verdure.’ But further on he made a second voyage to Surat, noid was able to oe longer there. _ Among the specimens he then collected was one which he termed Ricinoides malabarica a A specimen of this species was either given or lent urmann, who figured it as a Croton in 1768 (Flor. Ind. t. 63, fig. 1). In the same work ; fig. 1) Burmann figured, also as a Croton, another specimen of the same species, collected by Garcin in Java. Neither of these specimens is to be found now in the Burmann herbarium, a their whereabouts, if they still exist, cannot be traced. species, which is that now known as ‘Chrozophora Rottleri, A. Juss., will be dealt with more fully when Burmann’s drawings of these two specimens are discussed. History OF THE SPECIES UNDER CROTON. When Linnaeus published Croton tinctorium in 1753 (Sp. Pl. p. 1004) he possessed only one specimen of this species. That specimen, which came from BLOne pert is still in his _her- er cite the spevies collected by Gundel in Melos or the species from the Red Sea littoral attributed by Isnard to Lippi, or the species from India collected by Garcin. In 1774, however, 63 inability to distinguish a species with stellate-pubescent carpels from one with lepidote carpels has been that Koenig’s suitable name and excellent diagnosis were not published. It is possible that the younger Linnaeus was guided by the judgment of Burmann, who, in 1768, when publishing figures of two specimens of this particular species, both of them collected by Garcin, named one ot them Croton tinctorium (Fl. Ind. t. 62, fig. 1). It may be that Burmann was to some extent influenced in his decision to refer the second specimen of the same plant to Croton hastatum (l.c. t. 63, fig. 1) by the circumstance that the latter was collected at Surat, the former in Java. Later authors have at times been impressed by another difficulty when Batavia from seed collected at Surat in 1723, still so young when it was collected that it had not lost the characteristic first pair of leaves or developed the shape and lobing characteristic of the leaves of a fully grown plant; the figure of Croton hastatum, Burm. f., non Linn., represents the upper portion of a fully developed plant of the same species gathered at Surat in 1724. We do not know whether Garcin gave these specimens, or if he only lent them to Burmann. We may suspect that they were only lent, because they are not in the Burmann herbarium now. In that herbarium, however, there still are two specimens of this species. One of them was sent to Burmann as *‘ Croton arvense, Koen. Nelle-tschendi. Koenig n. 463.’ The other was collected by Sonnerat, also in South India. We know that officer in the service of the King of Denmark and before his transfer to that of the Nawab of Arcot. We know, too, that fter me year as his own, ‘bore the name Croton asper (or asperum), published for the first time, without description, in 1814 (Rorb. Hort. Beng. p. 104). It may be that when Burmann passed the proof of his figure of Croton hastatum, he was under the impression that the plant -depicted really was Croton hastatum, Linn. ‘The fact that the plate is so inscribed has led Miiller to remark (DC. Prodr. xv. 2, 64 mann may have thought when he passed the proof of his en- graving, we know that when the corresponding text was printed (Flor. Ind. p. 305 [205]) he regarded Ricinoides malabarica surattensis, Garcin, as differents from Croton hastatum, Linn.; e named Garcin’s plant ‘C. hastatum B.’ During his journeys in 1761-62 Forskal collected specimens of four forms of Chrozophora. One of the four was the ‘ Tourne- ‘sol’ itself; a specimen of this, written up b Forskal as Croton tinctorium, is now in the British (Natural History) Museum. The indication ‘ex Oriente’ has been added to the sheet by another hand. This specimen is not referred to in the ‘ Flora Aegypto-Arabica ’ edited by Zoega in 1775. No doubt the speci- men may have been collected in Egypt; the ‘ Tournesol ’ has often been found near Cairo. But Forskal also made a collection near Marseilles, and this British Museum specimen of Chrozophora tinctoria, A. Juss., is just as likely to have been gathered in the Bouches du Rhéne and omitted from Forskal’s Marseilles list as to have been obtained near Cairo and left out of the Egyptian one. The references by Zoega to the other three gatherings made by Forskal are not quite satisfactory. Yet as two of them have become the basis of still valid species of Chrozophora, it is necs- sary to examine them all with care. Two of the three are enu- . Ixxv.), as ‘490. Croton tinctorium’ and ‘491. Croton argen- teum’ respectively; the third is in the Arabian portion (l.c i “563. C species are as imperfect as that suggested for the Arabian one. The specimen of ‘491. Croton argenteum,’ duly written up im the Copenhagen herbarium, bears little resemblance to the true Croton argenteum, Linn., which is the South American plant now known as Julocroton argenteus, Didr. Nor is the treatment of ‘ 490. Croton tinctorium ’ much better. In the descriptive section of the ‘ Flora’ (Cent. vi. p. 162) caution as been enjoined; there the species is described as ‘ Croton tinctorium? * The type specimen, which was collected at Gizeh and is duly written up in the Copenhagen herbarium as ‘C. tinc- 65 torium Forsk. Cent. vi. p. 162’, has stellate- -pubescent capsules, and therefore cannot be Croton tinctorium, Linn., in which the capsules are lepidote. Forskal’s name Croton argentewm, which had no accompanying description, bears the relationship to the plant with which Forskal associated it, that Miiller imagined to subsist between Croton hastatum, 1 . non Linn., and Ricinoides mala- barica surattensis, Garcin. In this case, however, neither Miller nor Pax and Hoffman have ct ad that the epithet ‘argenteum ’ has priority. n 1786 Lamarck, under Croton, dealt in intention with only ign: in practice with three, species of Chrozophora. One of these was the Tournesol, C. tinctorium, Linn. (Encyc. Meth. iii. p- 214). In the Lamarck herbarium there is one specimen which Lamarck has written up:—‘ un rameau du Croton tinctorium. Tournesol d. Vahl.’ There is nothing to show whether this specimen had been given by Vahl to Lamarck before or after the publication of the third volume of the Encyclopedia. Mfrs the ‘ Tournesol ’ itself has been treated by Lamarck as it w Linnaeus, the treatment of 1753 has been departed from by ‘the addition of two varieties. The inclusion of these was the result of Burmann’s action in 1768. Lamarck appreciated, what Burmann failed to observe, that Croton tinctorium, Burm. f. non Linn., and C. hastatum, Burm. f. non Linn., are forms of the same species. But Lamarck, like Burmann, failed to observe that this species with stellate-pubescent capsules cannot be C. . tinctortum, Linn., in which the capsules are eee tc Lamarek regarded C. hastatum, Burm. f. non Linn., . tinctorium’ 8,’ citing Burm. Fl. Ind. t. 68, fig. 1 in the ‘text, and ye up a specimen in his herbarium, ven eae eres tly with that figure, as ‘ Croton hastatum Burm. Fl. Ind. t. 63, £.1. Croton tinctorium var. 3 enc.’ The locality and the collector of this specimen are not mentioned. Under ‘C. tinctorium y’ Lamarck in his text has cited Burm. Fl. Ind. t. 62, fig. 1, while in his herbarium there are four specimens of the same species, which hardly differ from the representative of his var. (3. These speci- mens Lamarck hire written up as ‘Croton tinctorium de Mr. Sonnerat. var. y enc.’ The other specins of Chrozophora described by Lamarck is vets senegalense (Encyc. Meth. iii. p. a: based by citation specimen referred to as ‘No. 165 A. dath.’ There is no exain pis of C. senegalense in the Lamarck gir barium. I[ Jussieu herbarium, however, we find evidence that while faceke phora to which he gave the field-numbers ‘ e am 60,’ Herb. de Galam 61,’ and ‘No. 145.’ The general facies of all three is the same, but whereas ‘Herb. de Gala t undersurface of the leaves softly woolly with long-stalked stellate hairs, the other two specimens, ‘ No. 145’ and ‘ Herb. de Galam 61,’ agree with each other, and differ from ‘ Herb. de Galam 60’ in having the leaves adpressed-hoary beneath with sessile ats hairs. Of these three. assis it is to ‘Herb. de B ? 66 with softly woolly leaves, that the original description of Croton senegalense, Lamk, applies. Yet it is one of the two with leaves adpressed-hoary on the undersurface which Lamarck has cited as his type, and the example of ‘ Adanson n. 145’ in the Jussieu herbarium has been written by Lamarck himself as ‘ Croton sene- galense Lamk encycl.’ The specimen which agrees with Lamarck’s description bears no endorsement in Lamarck’s hand- writing; beyond the words ‘Herb. de Galam 60’, written by Adamson, it only has the word ‘ Croton ’ written by A. L. Jussieu. The plants represented by this specimen and by ‘ Adanson n. 145” are certainly closely related. This, however, does not alter the fact that, under Croton senegalense, Lamarck has des- cribed one plant and cited another as the basis of his species. In 1790 Vahl (Symb. Bot. p. 78), dealt more satisfactorily than Zoega did in 1775 with the two species of Chrozophora collected by Forskil in Egypt. The one which in Forskals list stood as ‘490. Croton tinctorium,’ Vahl described as Croton plicatum, and the actual specimen at Copenhagen which bears the legend ‘©. tinctorium Forsk. Cent. vi. p. 162,’ has been written up by Vahl as C. plicatum. By some inadvertence, however, Vahl, when describing this species, either failed to note that it had been collected at Gizeh, or failed to realise that Gizeh is in Egypt. By giving ‘ Arabia,’ where C. plicatum has never yet been collected, as the home of the species, Vahl has led more than one subsequent author astray. The plant, which in F orskal’s that this error was not due to any confusion by Vahl between the plant collected at Gizeh which Forsk&él named C. tinctorium and the plant collected at Lohaja in Arabia to which Forskél gave the same name. It is true that Vahl did not take up ‘ 563. Croton tinctorium?’ in the ‘ Sym- belae.’ It is also true that one of the two examples of this gathering now in the Copenhagen herbarium, which originally formed part of the Schumacher herbarium, was not written up by Vahl. But the other example of the Lohaja plant at Copenhagen bears the name, in Vahl’s handwriting, ‘Croton tinctorium F.’ This endorsement shows that Vahl was aware of the treatment Forskal had- proposed for this plant, and suggests that Vahl realised that it is different from C. tinctoriwm, Linn. In 1794 the brothers Russell published a list of Aleppo plants (Aleppo ii. p. 265) one of these plants being Croton tinctorium, Linn. Their specimens, now in the British (Natural History) Museum, show that they included under the species at least one specimen which, while closely allied to, yet differs specifically from the true ‘ Tournesol ’ figured by Clusius and Gesner. * 67 Willdenow in 1805 included in Croton (Sp. Pl. iv. 1, pp. 538- 504) six species of Chrozophora:—20, tinctorium; 21, plicatum; 22, obliquum; 23, verbascifolium; 59, moluccanum; 67, sene- galense. Two of these, C. tinctorium and moluccanum were, in intention, taken up from Linnaeus (1753); one, C. senegalense, was taken up from Lamarck (1786); two more, C. plicatum and obliquum were taken up from Vahl (1790); the last, C. verbaset- folium, was based on Tournefort’s Ricinoides ex qua paratur Tournesol gallorum, folio oblongo et villoso (1703). In dealing with C. tinctorium (l.c. p. 538) Willdenow dis- carded the varieties proposed by Lamarck in 1786, but added to the Linnean citations of 1753 that of Heliotropium minus, as figured by Gesner, and as understood by Camerarius. In discussing C. plicatum (l.c. p. 538) Willdenow was less lio obl t villoso, 2 rair fecaing Munirnatort’s species afresh as C. verbascifolium (hes as Givotia rottleriformis, Griff. That species is not entitled to the epithet ‘molucanna’; the plant does not occ I Delica. We know that Linnaeus had no knowledge of ree B 68 by Willdenow (l.c. p. 590) as : cover (Herb. Willd. fol. 17891), however, has had written upon it by Willdenow the published diagnosis of Croton moluccanum with the additional note ‘habitat in India’ in substitution of the published habitat. The cover contains three specimens of the same species of Chrozophora. One of the three was sent by Roxburgh to Willdenow as C. asper; the other two were sent by Klein to Willdenow as C. plicatum. One of the two sent by Klein has smaller and less lobed leaves than the other. The description is brief but clear; it alludes to the variability of this species, more herbaceous in cultivated ground, more woody and with smaller leaves in waste places. The account of the capsules is exact. But Willdenow did not observe that this plant so well y varieties by Lamarck. This is the more singular, since his judg- ment as regards the specimens of the two different species sent to him by Klein under the name C. plicatum was so just. In dealing with C. senegalense, Willdenow (l.c. p. 5954) adopted Lamarck’s treatment, apparently without suspecting that Lamarck had described one plant and quoted another. herbarium (Geis. Crot. Monogr. pref. p. iv.) with the help of ahl’s specimens and notes. ti is, however, clear, from internal evidence, that Geiseler had no opportunity of consulting the speci- mens used by Burmann, Lamarck and Willdenow. Again there are six species that belong to Chrozophora: —70, senegalensis ; 87, Rottlert; 110, tinetorius; 111, viet 112, obliquus; 113, verbascifolius. In dealing with Croton senegalensis (l.c. p. 45) Geiseler has 69 In the case of C. Rottleri, Geiseler (i.c. p. 54) described as new a species of which a specimen had been sent to Vahl by Rottler trom Southern India. The species was not a new one because the specimen on which it was based represents the shrubby condition ot the plant described by Burmann in 1768 both as C. tinctorium, Burm. f. non Linn. and as C. hastatum j5, Burm. f. non C. has- tatum, Linn.; again described by Lamarck in 1786 both as C. tinetorium, var. j3 and as U. tinctorium, var. y; and described a third time by Willdenow in 1805 as C. moluccanum, Willd. non Linn. Since, however, none of these earlier names is valid, Geiseler’s name is that accepted for this species. In dealing with C. tinctorius, Geiseler (l.c. p. 68), by exclud- ing the reference, interpolated by Willdenow, to the figure by Gesner of a plant raised at Padua a century and a half earlier from seed received from Crete, reverted to the Linnean treatment of the lournesol in 1753. n the same specimen. The opportunity he had of studying that Geiseler’s citation. of the prostrate Indian species which is the basis of C. plicatum, Willd. non Vahl, was the result of his Lamk, was due to want of opportunity to examine the Lamarck herbarium. Had Geiseler seen the Sonnerat specimens in not to C. plicatum, Vahl, but to his own C. Rott ert. . In dealing with C. obliquus, Geiseler (l.c. p. 71) was again able, from an examination 0 the actual type, to supplement Vahl’s original description by adding that in this species the i his case the absence of an opportunity due to want of access to the Willdenow se of C. moluccanum, Willd. non L as described by Willdenow, includes the form described by Geiseler as C. Rottleri. Under 70 to accept as A. moluccana had been published as Jatropha moluccana, Linn. In 1812 Delile (Deser. Egypt. Hist. Nat. ii. Flor. Egypt. p. 139, t. 51, fig. 1) described and figured as Croton oblongifoliwm a plant collected by him at Ajeraud near Suez. This was the first valid name for a species which had already been obtained on two occasions; this is the plant in the Jussieu herbarium which Isnard a century earlier had wrongly identified with Lippi’s ‘ Ricinoides memphiticus folio laevi,’ and is also the species col- lected in 1762 at Lohaja in Yemen by Forsk4l in whose Arabian list it stands as ‘ 562. Croton tinctorium ? ’ In 1815 Sibthorp and Smith (Fl. Gr. Prodr. ii. p. 249) des- cribed as a new species, Croton villosus, the plant which in 1703 1790 as C. obliquum; and. Willdenow had again published in 1805 as Croton verbascifolium. The same form, when discovered for the first time in Spain, was published once more by La Gasca in 1816 (Gen. et Sp. Nov. p. 21) as Croton patulus. Burm. f. (1768) non Linn. Roxburgh’s belief that it was C. plicatum was derived from Willdenow, a circumstance which proves that Willdenow had not informed Roxburgh that the speci- men of C. asperum which he had sent to Willdenow had been treated as part of the basis of C. moluccanum, Willd. non Linn. a, prospectus issued by Sieber in 1821 there are three references under Croton to species of Chrozophora. These call for attention because they are very generally cited by later authors. The first (Avis des plantes, p.5) relates to Croton tinctorium, Linn. from Girapetro, in Crete. The specimens are not exactly like the ‘Tournesol’ of Southern France figured in 1557 by Clusius. They do, however, agree closely with the plant mentioned by Gesner in 1561 as having been raised at Padua from Cretan seed, and figured by him. The other two references (l.c. p- 7) are both to C. plicatum, Vahl, one of them noted in the Egyptian list, the other in the Palestine one. But the specimens show that the two plants are different; that from Egypt has stellate-pubescent capsules, and is nearly related to C. plicatum, Vahl; that fr Palestine, where it was gathered in the garden of tC tawaske: has lepidote capsules, and somewhat resembles, though it does not quite agree with, the Cretan plant cited as C. t L : ; inctorium. ter the issue of this prospectus, but before the distribution of the 71 corresponding specimens, Sieber discovered his error, and the printed labels issued with the plant from Gethsemane bore the name Croton oblongifolium. This was a new inadvertence, because C. oblongifolium, Del. does not occur in Palestine. But the use of the name C. oblongifolium, Sieb. (1821) non Del, (1812), is the first recognition of this plant as a distinct form, In 1824 appeared the brief but important summary by the younger Jussieu (Huphorb. Gen. Tent. pp. 27, 28, t. 7, fig. 25/1-11). This revindicated the generic status of Chrozophora but in so doing it only transferred by name the Tournesol itself, in connection with his plate, to Necker’s genus. Jussieu cited, as illustrating the genus, the various species of Croton which he believed to belong to Chrozophora, but left to those who might follow him the task of deciding what names these species ought to bear under Chrozophora. After defining the genus, Jussieu remarked.—Species circiter 8, duae senegalenses, caeterae ex Arabia aut Africa boreali quarum duae in Europa australi erescunt; inter quas scilicet Croton tinctorium, Linn.; C. obliquum, Vahl; C. plicatum, Vahl; C. verbascifolium, Willd. ; C. oblongifolium, Delile; C. senegalense, Lamk. Quibus ex descriptione forsan congeneres C, mollissimus, Geis. et C. Rott- leri, Geis., alter sinensis, alter indica. These carefully weighed statements of A. Jussieu can only be fully appreciated if the specimens in the Jussieu herbarium be examined. These specimens are arranged under seven consecutive numbers, 16262-16268. Herb. Jussieu 16262 includes two specimens, ‘ Adanson n. 165’ and ‘ Herb. de Galam 61.’ The two belong to the same form. The one marked ‘ Adanson n. 165’ is by citation the type of Croton senegalense, Lamk; it has been written up by Lamarck himself as ‘ Croton senegalense, Lamk encycl.’ and as been further inscribed by A. L. Jussieu (but not by A. Jussieu) as ‘ Crozophora senegalensis Ad. J. Euph. 28,’ although the entry on the page cited really is ‘ Croton senegalense, Lamk.’ Herb. Jussieu 16263 includes two specimens of Croton tine- torium, Linn., neither of them localised and neither with a collector’s name. Herb, Jussieu 16264 consists of one specimen only, without locality or collector’s name. Its chief interest is that it has been written up by Vahl, when he examined the Jussieu herbarium, as ‘Croton obliquum.’ ‘The specimen is exactly like the one at Copenhagen which was named Croton argenteum by Forskal, and was later made the type of Croton obliquum, Vahl (1790). Herb. Jussieu 16265 consists of one specimen only, which is interesting because neither A. L. Jussieu nor A. Jussieu, nor Lamarck nor Vahl when they examined the herbarium, ventured to name it. It originally belonged to Danty d’Isnard, who wrote it up as ‘ Ricinoides memphiticus folio laevi, Lippi.’ As has already been explained, a later note in an unknown handwriting has explained that it cannot be the plant Lippi had in mind. The specimen is one of Croton oblongifolium, Del. : Herb. Jussieu 16266 consists of one specimen, collected in Senegal by Adanson, whose note is :—‘ Herb. de Galam 60.’ This 72 ® is the specimen on which the description of Croton senegalense published by Lamarck was based, though it is not the specimen cited by Lamarck as the type of his species. Vahl, when he examined the Jussieu herbarium, discovered the transference which Lamarck had inadvertently made. ‘To rectify matters Vah! prepared a new description of Croton senegalense, based on ‘ Adan- son n. 165.’ This description, we have seen, he permitted Geiseler to publish in 1807. Neither A. L. Jussieu, A. Jussieu, Lamarck, nor Vahl gave this plant a specific name. It bore, when they studied it, only the word ‘Croton’; at a later date Desvaux added to the generic name the indication ‘ senegalense var. 3. Desv.’. The chiet interest of the specimen is that it was the basis of the second of the ‘ species duae senegalenses’ referred to by A. Jussieu. Herb. Jussieu 16267 consists of two specimens, rather badly prepared, of the herbaceous condition of Croton Rottleri, Geis. The chief interest of these specimens is that they. have been written up by Vahl himself, as Baillon has already stated (H'tud. gén. Euphorb. p. 322), as ‘Croton plicatum, Vahl.’ To this indi- cation A. L. Jussieu has added the word ‘Symb.’ and has at the same time written ‘ Crozophora plicata, Ad. J. Euph. 28.’ A secondary interest attaches to this evidence that Vahl, when he mens were gathered in Southern India. Even this information is indirect. The writing on the labels is in Tamil, but in both cases the inscriptions, which contain the name ‘ Rama,’ are apparently catchwords from the edge of some manuscript. Herb. Jussieu 16268 consists of two specimens, both representing the same condition of the same species. One of the specimens bears three labels written up by Gundel, Tournefort, and Vaillant respectively. The label by Gundel is endorsed ‘ex insula Melo.’ That written up by Tournefort reads, ‘ Ricinoides ex qua paratur Tournesol gallorum folio oblongo et villoso,’ so that we are left free from doubt as to the plant intended by that name as published in 1703 (Cor. p. 45). The second specimen bears two labels, one of these endorsed ‘Croton verbascifolium, Willd.’ It is possible that the specimen may have been received from Willdenow him- self, for underneath the original name A. L. Jussieu has written ‘Crozophora verbascifolia Ad. J. Euph. 28.’ The second label, however, is more interesting; it is in the handwriting of Adrien Jussieu himself, who has endorsed it ‘Croton obliquum,’ an identification which, under the circumstances, was fully justified, because the specimen agrees well with the one in Herb. Jussieu 16264, named Croton obliquum by Vahl himelf. This endorse- ment perhaps explains the younger Jussieu’s statement, not so fully appreciated as it deserves, that two of the North African species of Chrozophora extend to Europe. One of the two is the ‘Tournesol ’ itself, which is to be met with everywhere on or near the Mediterranean seaboard; the other is Croton obliquum, Vahl, which, until A. Jussieu wrote, had been looked upon as 73 exclusively African, the name C. verbascifolium, Willd., having become established for the same species when collected in Europe. As regards the two species which he only knew from description, Jussieu was not so fortunate. The Chinese one, Croton mollissi- Croton plicatum, Vahl, there being no example of that prostrate African species in the Jussieu herbarium. Finally, though that herbarium did contain an example of Croton oblongifolium, Del., Jussieu did not realise the fact. His citation of that species was based upon Delile’s description and figure. In 1830 the task of distributing the specimens of Chrozophora which had accumulated in the East India House was carried out by Wallich. These specimens were aggregated under one entry (Wallich, Cat. Lith. 7716) subdivided into nine sections (A.-I.). Though less important than the corresponding specimens in the Jussieu herbarium because they never became the basis of a literary contribution to the study of the genus, they are nevertheless of some consequence owing to the frequency with which they have been cited. For this reason it is deeingbte to explain precisely what they are. ‘77716 A. Croton tinctorium, Herb. Roxb.’ The name of this specimen was written up by Roxburgh. The specimen represents the herbaceous condition of Croton Rottleri, Geis. ‘7716 B. Herb. Heyne.’ This includes two specimens, with neither name nor locality. One is Croton Rottleri, Geis., and purple ripe capsules. As regards its capsules it agrees with the asis of C. plicatum, Willd. non Vahl, but it differs from the Tiruvalur plant in having the leaves eglandular at the base and particular form, w or Vahl, or Geiseler, or Willdenow. ‘7716 C. Crfoton] plicatum e Patna, et Crfoton] asperum e amilt.? These two specimens were collected ta rt, i This manuscript was not published by the East India | the less, the 74 mature conclusion of one of the ablest of Indian botanists. The plant named C. plicatum, Ham., is the same as the prostrate plant callected by Heyne in Mysore in 1800; that named C. asperum is the herbaceous condition of C. Rottleri, Geis. his Indian career, which opened in 1819. They represent the same two plants distributed under 7716 C. The Pierwandy plant is C, plicatum, Ham. non Vahl, nec Willd., nec Roxb. The Madras plant is the herbaceous state of C. Rottleri, Geis. ‘7716 E. Crfoton] plicatum, Herb. Madr.’ Neither locality nor collector’s name accompany this specimen. We know, how- ever, from other sources that many of the specimens belonging to the ‘Madras’ herbarium’ were collected by Rottler. This particular specimen is the shrubby condition of Croton Rottleri, Geis., and may well be a cotype of that species. If this be not the case, the origin of the plant sent by Rottler to Vahl cannot be traced with precision. It does not, however, follow that it who had gathered it at Devanur. It is therefore not impossible that Wallich n. ‘7716 E. Herb. Madras’ may also have been gathered by Klein. ‘7716 F, Crfoton] asperum Herb. Russ.’ This specimen was collected by Dr. Patrick Russell, F.R.S., and therefore, as we ow from other sources, came from the Northern Cirears. The plant is the herbaceous condition of C. Rottleri, Geis. ‘7716 G. Hindustania.’ This specimen has been cited by Miller (DC. Prodr. xv. 2. p. 749) as having been collected by Wallich. Other authors have copied Miller. The label with the indication ‘ Hindustania’ was, however, written by Roxburgh. The field-note, on the other hand, was written by Hardwicke, and indicates that the plant was collected at Bindraban, a sacred city on the Jumna close to Mattra. It came, therefore, from a locality which neither Roxburgh nor Wallich ever visited, and one to which neither of these botanists ever sent their own collectors. The chief interest of this plant is that it is the first Chrozophora _ with lepidote in place of stellate-pubescent capsules to be found in India. Though closely allied to the ‘Tournesol,’ Croton tinc- 75 Larumng Linn., it is not identical with the ‘ Tournesol, but agrees hand to the herbaceous condition of Croton Rottleri Geis. ‘7716 H. Cr[oton] polycarpum, H. B. Cale.” This also repre- sents the annual herbaceous condition of C. Rottleri, Geis., which is so beak e in Bengal. The authority for this particular name is unknow 4710.1. paced Mew, 1826.’ This specimen again rioresente ne herbaceous condition of Croton Rottleri, Geis. It was collected by Wallich himself at Paganmyo in Burma. It is not, however, the first record of the species from that country. There is a speci- men in the British (Natural History) Museum which was collected at Kyouk Zeik in 1795 by Buchanan (afterwards ee when that officer was attached to the Embassy under Symes to the Burmese Court at Ava. History oF THE Species, 1826-64. While the genus Chrozophora was recognised anew by A. Jussieu in 1824, his treatment of its species was rather the last chapter of their history under Croton. Their history under Necker's genus opened in 1826 with a résumé by Sprengel yt: Veg. ili.) of ten species, eight of them attributed to Jussieu 1. Chrozophora senegalensis, A. Juss. ex Spr., is Croton senegalense as treated by Geiseler in 1807. Z S Rottleri, A. Juss. ex Spr., is Croton Rottlert, Geis. (1807). C. Sas itt A, Juss, ex Spr., is Croton oblongifolium, Del ot 2). 4, C, plicata, A. Juss. ex Spr., includes Croton ae eee Vahl a, and Centon plicatum, Willd. (1805) non V: C. hieroselymitana, Spr., is Cro ton oblongifolium, Sieb. ssi) i Del., now for the first ‘iene provided with a name that Ne ie asia. A. Juss. (1824), is the ‘ Tournesol.’ 7. C. obliqua, A. Juss. ex Spr., is Croton obliquum, Sieb. (1821). 8. C. verbascifolia, A. Juss. ex Spr., is Croton verbascifolium, bbongee (1805). C. mollissima, A. Juss. ex Spr., is bas mectieeterat, Geis. (1867) which is Mallotus perenne 2 Miill.-a 10. C. Burmanni, Spr. is Croton hastatum B. Burm. f. (1768). 76 In this list, therefore, we find evidence that Sprengel had not ascertained that C. obliqua and C. verbascifolia are merely different names for the same species, or learned that C. Burmanni is merely a somewhat different condition of C. Rottlert. In 1834 Decaisne (Flor. Sin. p. 242) recorded C. oblongifolia, A. Juss. as collected by Bové in Sinai, an important indication, since the identification is accurate, that as yet no confusion had arisen with regard to the name of Delile’s plant. We have already discussed the considerations which may have led Visiani in 1836 to refer Croton obliquifolium (Pl. quaed. Nilot. p. 39, t. 7, fig. 2) and Chrozophora Brocchiana (1.c. t. 8, fig. 2) to different genera. Dealing with them as individual species, Visian1 suggested that the former might be the same as Croton tinc- torium?, Forsk. (Cent. vi. p. 162); it certainly is very near that plant, which is the type of Croton plicatum, Vahl (1790). Pro- fessor Saccardo informs us that there is not now at Padua any specimen written up by Visiani as ‘Croton obliquifolium,’ and the present whereabouts of Visiani’s type cannot be traced. The same, Professor Saccardo has explained, is also unfortunately true as regards Visiani’s second species, the description of which does not agree exactly with the corresponding figure. The plant described is clearly closely allied to the one described by Lamarck s Croton senegalense, but differs in having leaves softly woolly above as well as beneath, whereas in Lamarck’s plant the leaves are nearly and at length quite glabrous above. The relationship of the various forms in the group Senegalenses will be dealt with in a subsequent paragraph. 3 In 1839 Fischer and Meyer (Kar. Enum. Turc. p. 171) named, without description, a new species, C. gracilis, from Turkestan. In 1841 Karelin and Kirilow (Bull. Soc. Nat. Mosc. xv. p. 446) described a new species, C. sabulosa, from Soongaria. In 1844 Presl (Bot. Bemerk. p. 109) described as C. Sieberi the plant that Sprengel in 1826 had nam . hierosolymitana. About 1850 Ledebour dealt with all three (Fl. Ross. i. 2. p. 581) keeping C. gracilis and C. sabulosa, which are identical, as distinct species, but merging Caucasus specimens of C. hierosolymitana in C. tinctoria, A. Juss. In 1851 Bunge (Rel. Lehm. p. 314) recognised C. sabulosa but not C. gracilis, and (l.c. p. 315) treated Turkestan examples of Sieber’s plant as pessibly distinct from C. tinctoria. At the same time Bunge re-described as a new species, C. integrifolia, what Sprengel had diagnosed as C. obliqua, A. Juss.” In 1851, also, Richard (Tent. FI. Abyss. ii. p. 252) recorded C. plicata, A. Juss., from Abyssinia; all the specimens he dealt with are identical with Croton plicatum, Vahl (1790). But Richard collected in Lower Egypt specimens, which he identified with C. plicata, that agree with Croton obliquifolium, Vis. These specimens prove that as late as 1851 no author had regarded the plants of Vahl and Visiani as distinct. ,In 1858 Baillon issued a list of nine species of Chrozophora (Etud. gén. Euphorb. p. 322). This list is rarely studied, perhaps because it has no descriptions. Under each of his species, how- 17 ever, Baillon has cited representative specimens which are more reliable than descriptions can possibly be. These species are : — 1. Chozophora integrifolia, Bunge, based on Lehmann’s plant described by Bunge. 2. C. sabulosa, Kar. et Kir., based on Lehmann’s plant cited by Bunge. 3. C. tinctoria, A. Juss., the ‘ Tournesol.’ 4. C. senegalensis, A. Juss., which includes ‘ Adanson n. 165’ the plant cited by Lamarck as the type of Croton senegalense; the plant collected by Adanson at Galam on which the description of Croton senegalense was based; and a specimen, ‘ Kotschy n. 28,’ from Abu Gerad in Kordofan which agrees with ‘ Adanson n. 165’ as regards tomentum on the under-surface of the leaf, but differs from both the Senegal plants of Adanson in having the leaves persistently shortly puberulous above. 5. C. obliqua, A. Juss., based on the plant in Herb. Jussieu written up by Vahl as Croton obliquum. 6. C. oblongifolia, A. Juss., represented by specimens collected by Bové and by Botta in Sinai, which agree with Croton oblongi- folium, Del. (1812). 7. C. verbascifolia, A. Juss., which includes Croton verbasci- folium, Willd. and, with this, three specimens collected b Aucher (2008, 5297 from Persia: 2006 from Syria), all of them belonging to C. hierosolymitana, Spr., as does another unlocalised specimen collected by Botta. r specimen without field note, which agrees with Croton obliqui- folium, attributed on the herbarium label to Olivier: and 9. C. obliquifolia [Vis.], Baill., based on a specimen of * Kot- schy n. 473,’ from Wolet Medine in Sennar, which agrees with Visiani’s figure of Croton obliquifolium. i i In 1860 T. Anderson (Fl. Aden. p. 36) recorded C. obliquifolia from the coasts of the Red Sea and the Gulf of Oman. This record, which Anderson’s specimens prove to be correct, is of interest as showing that at t is late date no confusion had arisen as to the proper name of the species. In 1860 also, Baillon (Adans. i. p. 67) reverted to the forms from Senegal, recognising two species, one of them C. senegalensis, A. Juss., treated as Geisele treated Croton senegalense in 1807; the other C. plicata, A. Juss., 78 he did not cite. As the Chrozophora to which the specimen be- longs has lepidote capsules, whereas the capsules ot C. licata are stellate-pubescent, this reduction proposed by Baillon cannot be adopted. In 1861 Dalzell (Fl. Bomb. p. 233) carried out what Hamilton had proposed in 1822 and treated the erect Indian Chrozophora with stellate-pubescent red-purple capsules as distinct from the prostrate Indian one with eglandular leaves and blackish non- tinctorial stellate-pubescent capsules. But whereas Hamilton termed the erect species Croton asperum and the prostrate one Croton plicatum, VDalzell termed the erect species Chrozophora plicata and used for the prostrate one the new name Ch rozophora prostrata. In 1862 Schweinfurth (Pl. quaed. Nilot. pp. 10, 11, tt. 3, 4) carefully figured and separated the two forms described as Croton plicatum by Vahl in 1790 and as Croton obliquifolium by Visiani l 36. In doing this Schweinfurth, after hosie examined the type of ‘Croton tinctorium? Forsk. Cent. vi. p. 162’ in the Copenhagen herbarium, rejected Visiani’s suggestion that Croton obliquifolium, Vis., might be the form described by Forskal. Nevertheless, after having taken this step, Schweinfurth used for Visiani’s plant the name Chrozophora plicata, in spite of the fact that the type of Croton plicatum, Vahl (1790), is the plant from Gizeh which Forskal had described as Croton tinctorium. For the plant with stellate-pubescent capsules which does agree with ‘ Croton tinctorium? Forsk. Cent. vi. p. 162° Schwein- furth used the name Chrozophora obliqua, notwithstanding the fact that in 1807 Geiseler had explained that the capsules of Croton obliquum, Vahl (1790) are lepidote. Under the plant which Visiani had termed Croton obliquifolium in 1836 and which Baillon had named Chrozophora obliquifolia in 1858, but which Schweinfurth had now termed Chrozophora plicata, the last-named author in 1862 cited a specimen which Klotasch in 1861 (Peters, Mossamb. Bot. ii. p. 99) had identified with C. tinctoria. In 1863 Klotzsch (l.c. corrig. p. 576) ac- cepted Schweinfurth’s emendation. This plant, Schweinfurth has remarked, does not extend to India. But the plant which Schweinfurth had now termed Chrozophora obliqua was treated with less caution because it was made to in- clude, in addition to the African form which agrees with the type of Croton plicatum, Vahl (1790), both the South India Croton plicatum, Willd. (1805), non Vahl, for which, Schweinfurth has told us, Klotzsch had suggested the name Chrozophora parvifolia, and the Bombay plant which Dalzell had named Chrozophora a variety Hartmanniana (t. 5) one which better agrees with 79 Visiani’s description. He also suggested that the Senegal plant with leaves woolly pei with long hairs might be another variety of C. Brocchia In 1864 Thwaites (teed: Pl. Zeyl. p. 443) first, and rb ne among authors who had dealt with Chrozophora since 1807, used the name C. Rottleri for the plant described by Geiseler as sites Rottleri, without confusing that plant with any other species. CHrozopHorsa IN DeCanpoLLE’s PRopROMUS. The important monograph of the genus by Miiller in 1866 (DC. Prodr. xv. 2) has already been adverted to. The relation- ship to each other of the various forms recognised could hardly f : eclectic nomenclature are, however, less satisfactory. This is not because of the extreme reduction advocated; although Miiller included all the forms of the Plicatae group in one species and all those of the Tinctoriae group in two species, he did not take advantage of this to shirk any of the issues involved; he care- fully treated as distinct varieties the forms which earlier writers had regarded as separate species. Thus C. plicata, ae -arg. (l.c. p. 747) includes three varieties :—a. Rottler:, 3. genur y. prostrata; while C. tinctoria, Miill.-arg. (l.c. p. 748) facliades four raeeiete ge verbascifolia, (3. hierosolymitana, 7: genuina, and 6. subplic Bist: ag a. Rottleri is treated naturally. It includes | Croton Rottlerr Geis. (1807); Croton hastatum 8. Burm. f. (1768), which is the basis of Chrozophora Burmanni, Spr. (1826) ; Croton moluccanum, Willd. (1805), non Linn. It also includes — tinctorium, ie xb. ex] Wall. (1830); Croton a . ( a. Rottleri the Java Croton tinctorium, Burm. f. (1768) non ct which is only a cultivated condition of Croton hastatum o. 2 has excluded Croton asperum, [Ham. ex] Wall. (1830), ‘hich i is the same thing as Croton asper, Koen. (1814); he has also ex- cluded Croton plicatum, Roxb. (1814) non Vahl, which, as Rox- burgh explained in 1832, is the same thing as Croton tinctorium, Burm. f. on the one hand and Croton asperum, Koen. on the other. More perplexing still is the reference by Miiller of Croton asperum, Ham., which by the actual type is his a. Rot- tleri, to his (3. genuina; and the reference of ng plicatum, Roxb., which also by its co-types is Miiller’s a. Rottleri, to his - prostrata. Misled by the employment by Wailisnow of the epithet ‘moluccanum,’ Miller has attributed a plant sent by Koenig from Devanur in Madras to the Moluccas, where no Cc hrozophora occurs. 1b. C. plicata 8. genuina, believed by Miiller to correspond with Croton plicatum, Vahl (1790) in reality carefully excludes Vahl’s plant, since the variety has been based upon Chrozophora plicata, Schweinf. ( 1862) which is Hains oblauslol mang: This variety, copying Schweinfurth, includes C. , Klotzsch 80 (1861) non A. Juss. But Miiller did not accept the conclusion of Schwerntuith that this African Chrozophora is absent from India, or accept the opinion of Schweinfurth that both of the prostrate Indian forms of this genus in which the capsules are stellate- pubescent belong to the same variety of Croton plicatum, Vahl. The former decision was less, the latter was more satisfactory than the conclusions of Schw einfurth, which they respectively contra~ dict. What, however, is most perplexing in Miiller’s action is that, of the two prostrate forms of OR recognised by him, the specimens cited show that it is the one which Dalzell described as C. prostrata that Miller a referred to his variety Se Aiea’ Vahl, whieh is Cheskenhirs pes Klotzsch ex Schw ein Ha tg, . genuina ‘By some inadvertence the cate: on which C. parvifolia, Klotzsch, was based has been attributed to Malacca, whence no Chrozophora has yet been reported, although Klein really collected it at Tiruvalur, near Madras. A corollary to the reduction to his var. y. prostrata of Croton plicatum Roxb., has been Miiller’s impossible reference to this particular form of ‘roton tinctorium, Burm. f., which was raised in a Java garden from. Surat seed. 2. C. sabulosa (\.c. p. 748) is the plant described under this name by Karelin and Kirilow in 1842 and again by Bunge in 1851. The existence of C. gracilis, named but not described by Fischer and Meyer in 1839, and described as distinct from C. sabulosa by Ledebour in 1850, was not alluded to by Miiller. 3. C. tinctoria a. verbascifolia (l.c. p. 748) is naturally de- limited end includes Croton Sinctene Willd. vere ; 3b. C. ae ene ( €. p. ere is bein on the specimen which is the type equally of Croton oblongifolium, Sieb. (1821) non Del., of Phrcsaihons A ne Spr. (1826), and of C rozophora Siebert, Pres] (1844). There is no confusion between this plant and a. verbascifolia; indeed it is difficult, not- Cairn 4 what ms been said and done by later Wuthors. to think of such a confusion as possible. But there was some con- fusion between 23. hierosolymitana and y. genuina, to which latter 81 plant the Spanish and the Greek specimens cited under var. (3. belong. ; dc. C. tinctoria y. genuina (l.c. p. 749) is confined to the ‘ Tour- nesol,’ as figured by Clusius in 1557 and as described by Linnaeus, under the name Croton tinctorium, in 1753. This plant, however, occurs in two states, readily distinguished by the shape of the leaves. The plant from Southern France has these rather rhom- boid in outline, as shown in Clusius’ figure; many of the speci- mens from Greece, Crete, Asia Minor and Syria have the leaves ovate and subcordate, as shown in a figure prepared by Gesner about 1561, first published by Camerarius in 1586. This latter state of the ‘Toursenol’ outwardly much resembles some speci- mens of var. (3. hierosolymitana, to which, notwithstanding differences in the flower, Miiller has referred it. 3d. C. tinctoria 8. subplicata (l.c. p. 749), is a form which agrees with var. y. genuina in all its characters save that it is of a pros- trate in place of an erect habit. . C. obliqua (l.c. p. 749) is in intention Croton oblongifolium, Del. (1812), which is, as Miiller has stated, identical with Croton tinctorium? Forsk. (1775) from Arabia, as opposed to the Croton tinctorium? Forsk. (Cent. vi. p. 162) from Gizeh in Egypt. But Cc 82 Since the appearance of Miiller’s Monograph in 1866 the atten- tion given to Chrozophora has been mainly confined to the forms met with in Africa; to a less extent attention has been bestowed on those which occur in India or in the Orient. Nothing new has had to be said or suggested with regard to those forms met with in the Occident. History oF THE AFRICAN Species, 1767-1912. In 1867 Schweinfurth enumerated four African forms (Beztr. became apparent. Schweinfurth accepted Miiller’s view that C. obliqua, Schweinf. (1862) SI be C. obliqua, A. Juss. (1826), which is - certainly sound. But he also accepted Miiller’s decision that C. obliqua, Schweinf. (1863) 3 is sasties with C. prostrata, Dalz. noe tir A. Juss., a decision Sahie an examination of Croton argenteum, Forsk. non Linn., the basis of Croton obliquum, ahl, proves to be without justification. Still further on in the same treatise (l.c. Aufzahl. p. 262) Schweinfurth supplied a new list which includes six African names. Of these . C. Brocchiana, from Nubia, is the plant figured by Schwein- furth under this name in 1862. 2. C. obliqua, from the coasts of Egypt and Nubia, which is C. obliqua, Mill.-arg. (1866) and not C. obliqua, Schweinf. (1862) of the avkion list (l.c. p. 35). 3. C. plicata from Egypt, Nubia, Sennar, “and Bahr-el-Abiad, which is still C. plicata, Schweinf. (1862) and therefore is Croton watt 790.” Vis. (1836), as contrasted with Croton plicatum, a 4. C. prostrata, from Egypt, Nubia, paren. Sennar, and (1861) is ‘not that Indian plant, but is Pik African portion of C. plicata y. prostrata, Mill.-arg. (I866) and therefore at the same time is C. obliqua, Schweinf. (1862), non a Juss. (1826) nec Mill.-arg (1866). It is the true C. plicata, A. Juss. (1826), because it is the plant described by Vahl in 1790 as Croton get oe negalensis, from Kordofan, is the plant collected by Kotoake a Abu Gerad (Kotschy n. 25) which Baillon in 1858 had included in C. senegalensis, A. Juss., notwithstanding the fact that in this plant the homomorphic leaves are permanently ane scent an e green above, whereas in Croton sene galense, as cited by Lamarck in 1786 and described by Vahl in Gonaler 8 83 monograph of 1807, the dimorphic leaves are dark green and glabrous above. 6. C. tincteria, from’ Egypt, Nubia, and Abyssinia, includes two distinct species. The HKgyptian plant is the ‘ Tournesol ’ itself, Chrozophora tinctoria, A. Juss. (1824). That from the coasts of Nubia and Abyssinia is not the ‘Tournesol’ with red- purple ripe capsules, but the condition assumed in its first year by the species with blue-purple ripe capsules, which is C. oblongi- jolia, A. Juss., but which Schweinfurth, copying from Miller, had been misled into believing ought to bear the name C. obliqua. The account of the genus by Boissier in 1879 (Fl. Orient. iv.) includes two forms which are characteristic only of that part of the Orient which lies within the African continent, viz. :—Chrozo- phora plicata and C. tinctoria 8. subplicata, though all save one of the other Oriental species also occur in Africa. The interest which attaches to Boissier’s treatment of C. plicata lies in the fact that this species is held,both by the references and the speci- mens cited, to include Croton plicatum, Vahl (1790) and Croton oblongifolium, Vis. (1836). These two plants, considered by Boissier to be indistinguishable, were regarded by chweinfurth in 1862 as separable species and by Miller in 1866 as separable varieties. Schweinfurth, in 1867, while accepting Miller’s views as regards the names the two should bear, did not adopt Miiller’s decision that they should be regarded as conspecific. The interest which attaches to Boissier’s treatment of C. tine- toria 8. subplicata lies in his suggestion that this form may be a natural hybrid between C. tinctoria and C. plicata. It has to be said, in favour of this view, that C. subplicata is confined to the small area within which the regions occupied. by C. tinctoria and But apart from the circumstance that C. sub- plicata has the prostrate habit characteristic of C. plicata there is nothing to justify the suggestion. In every other feature C. characters of Miiller’s proposed variety. Boissier had observed, se A. Juss. (1826), a view whic _Schweinfurth ‘ . 2, p. 306). Ascherson and Schweinfurth in 1887 (l.c.) also for the first time recorded C. tinctoria, var. hierosolymitana, Mill.-arg., as an Egyptian plant. es 84 In 1888 Balfour dealt with the forms of Chrozophora trom Socotra (Trans. Roy. Soc. Edin. xxxi. p. 277) and enumerated three: —Chrozophora tinctoria, which is the condition during its first season of C. oblongifolia, A. Juss. (1826) and therefore is C. tinctoria, Schweinf. (1867) non A. Juss.; C. obliqua, which is normal C. oblongifolia, A. Juss. (1826) and therefore is C. obliqua, Miill.-arg. (1866) non A. Juss; lastly C. obliqua var. frutescens, Schweinf., which is based on specimens of C. oblongi- folia, A. Juss., from a plant older and more woody than usual. The species to which all three Socotra forms belong is rather variable in habit and appearance, a fact which is emphasised by the circumstance that in 1899 Schweinfurth (Bull. Herb. Boiss. vii. App. 2, p. 306) proposed the recognition of two more varieties, angustifolia and incisa, neither of which represents more than a condition of the type. In 1906 Broun enumerated four forms of Chrozophora from the Sudan (Cat. Sud. Fl. Pl. p. 72). These are C. obliqua from the Red Sea coast at Suakin which is C. obliqua, Mill.-arg. (1866) non A. Juss. and therefore is C. oblongifolia; C. Brocchiana from Dongola, Khartum, Sennar and Kordofan, supposed to be the ‘species of Visiani as understood by Schweinfurth in 1862; C. senegalensis, based on the plant collected by Kotschy at Abu Gerad in Kordofan, an identification accepted from Miiller’s monograph of 1866; lastly C. plicata, the plant common “ on river-banks and in depressions on cotton-soil in most parts cf the Sudan,’’ which was the basis in 1790 of Croton plicatum, Vahl. In 1912 an official account of the tropical African forms was published by Kew (Fl. Trop. Afr. vi. 1). In this account four species were recognised, viz. :— 1. Chrozophora plicata, A. Juss. (1826), delimited as it was in 1887 by Ascherson and Schweinfurth with the subdivision adopted by those authors into two varieties; one corresponding with Croton plicatum, Vahl (1790), the other corresponding with Croton obliquifolium, Vis. (1836) which is also Chrozophora obliquifolia, Baill. (1858) 2. C. oblongifolia, A. Juss. (1826), under the name which it had borne without question until in 1866 Miiller inadvertently mis- identified the species with Croton obliquum, Vahl (1790). 3. C. senegalensis, A. Juss. (1826), as the equivalentof Croton senegalense as cited by Lamarck in 1786 but as described by Vahl, through the agency of Geiseler, in 1807; with at the same time a variety lanigera, based upon the plant described as Croton sene- galense by Lamarck in 1786 but not cited by that author. 4. C. Brocchiana, Vis. (1836), also with two varieties, one of which is that figr age Schweinfurth as var. Hartmanni in 1862, ich a form for w renberg had proposed the name Croton macrocalyz. History oF THE Inpran Species, 1869-1906. In 1869 Stewart alluded to the species of Chrozophora of the Panjab (Punjab Pl. p. 192). The influence of Miiller’s mono- _ graph on his treatment is evident. Stewart recognised only two 85 species, C. plicata and C. tinctoria. The former is C. plicata, Mill.-arg. (1866) in the wide sense in which all the forms with stellate-pubescent but not lepidote capsules are regarded as con- specific; it therefore includes the erect form with long racemes and purple capsules as well as the prostrate forms with condensed racemes and non-tinctorial fruits. The latter, however, is treated more critically, because Stewart has explained that the plant in- tended is not any of the four varieties of C. tinctoria segregated by Miller, but is the Indian plant issued by Wallich which Miiller referred to C. obliqua. It is further noticeable that Stewart, by citing this as C. oblongifolia, indicated that he had already appreciated the fact that Miiller’s identification of C. oblongifolia with Croton obliquum, Vahl, could not be accepted. In 1870 King (Pl. N. W. Prov. p. 15) accepted Stewart’s treat- ment. So, too, did Aitchison with regard to the species with lepidote capsules which he identified with C. tinctoria in 1880 (Trans. Linn. Soc. Bot. n.s. vii. p. 108) and again in 1881 (Journ. Linn. Soc. xix. p. 186). The C. tinctoria of all three authors is not, however, the true ‘Tournesol’ of Europe but is C. hiero- sclymitana, Spr. (1826). é In 1887 Hooker supplied a critical account of the Indian species (Fl. Brit. Ind. v.) in which he has added a third species, C. ebliqua, to the two recognised by Stewart and King. prehensive ‘species’ into respond respectively and therefore with orm wh (1862); (c) Chrozophora prostrata, Dalz. (1861). In 1906 Cooke supplied yet another critical revision of the forms from Western India (Fl. Pres. Bomb. ii.) in which the number of species is now increased to four; C. tinctoria, C. obliqua, C. plicata and C. prostrata. . C. tinctoria, Cooke, i ribed as having the capsules both sth, ge bet: sd " stellate-pubescent and lepidote, which is not the case in 86 species of Chrozophora. No Chrozophora with lepidote capsules has ever been met with in the presidency of Bombay to the south of Scinde: The specimens cited by Cooke all have stellate- pubescent capsules. Therefore C. tinctoria, Cooke (1906), non . Juss., is precisely equivalent to Uroton tinctorium, Burm. f. (1768), n n Linn. a cbligua Cooke, is exactly equivalent to U. obliqua, Mill.- arg. (1866), non A. Juss., and therefore is in reality C. oblongi- folia, A. are (1826). 3. C. plicata, Cooke, is exactly equivalent to Croton plicatum, Roxb. dsi4),. non Vahl, and therefore is in reality C. Rottleri, A. Juss. (1826). 4. C, prevent, Cooke, is in intention equivalent to C. prostrata, Dalz. (1861). In practice, however, it also includes C. parvifolia, Klotzsch (1862), and that the limitation of the species was still further misunderstood we know from the circumstance that Cooke has included, by citation, under Dalzell’s species, the African C. plicata, A. Juss. (1826) which is Croton plicatum, Vahl (1790). History oF THE Orrentat Specres, 1879-1915. When in 1879 Boissier described the species of Chrozophora from the Orient (Fl. Orient iv.) he dealt with six different forms: —Chrozophora plicata, C. gracilis, C. tinctoria, C. tinctoria var. subplicata, C. verbascifolia, C. obliqua 1. C. plicata, Boissier’s specimens of which came tery oo African part of his area, has already been he ed. The tions under this African species of C. Rottleri, A. Juss. (1826) and C. prostrata, Dalz. (1861), neither of whith occurs in the ‘ Orient,’ suggest that the species was not fully understood. 2. C. gracilis, from Central Asia, is the plant so named by Fischer and Meyer in 1839 and described under their name by Ledebour in 1850. But this species is identical with C. sabulosa, named and described by Karelin and Kirilow in 1842 and described again under their name by Ledebour in 1850 and by Miller in 1886. Miller thus used the name with which a description was first associated, Boissier used the name which was first applied, without description, to this plant. 3. C. tinctoria, by the specimens cited, utes besides those that are referable to Croton tinctorium, Linn. (1753) some that are really referable to Chrozophora Sept Berne eine Spr. (1826 3b. C. tinctoria var. subplicata, which is confined to Lower Egypt, has already been discussed. 4. C. verbascifolia is no longer the natural species named Croton verbascifolium by phe cade - 1805 but is eaicresamng to be " C. “pbli ua, as Selcnzied by Boissier, was meant to be ¢. obliqua, Mill.-arg. and therefore to be identical with. Croton oblongifolium, Del. (1812). In practice, however, C. obliqua, 4 ae eee 87 Boiss. was made to include specimens from Wadi Dachel in the Libyan waste, written up by Ascherson as C. obliqua, which really belong to the PeTPaaD species described by Vahl in 1790 as Croton obliquum In 1915 an instance was afforded of the difficulty experienced in correcting an inadvertence which has become incorporated in an authoritative work. In a careful cies of the vegetation of Aden, Blatter (Rec. Bot. Surv. Ind. vii. p. 332) has taken up from Miller not only the erroneous vetdsdhie to an Indian plant issued by Wallich (Cat. Lith. 7716 G) but also the erroneous name Chrozophora obliqua, Mill.-arg. (1866), non A. Juss. The adop- tion of Miiller’s name is in this instance the more remarkable when regard is had to the circumstance a the plant had its true name roe wal ti by sie and Hoffmann in 1912, and that a 1860, di mn the true name, C. smaft, 8 Tue SPECIES IN ENGLER’S PFLANZENREICH. The treatment of the genus in the ‘ Pflanzenreich’ in 1912 has already been explained. The species recognised are nine in number; one of the nine it has, however, been suggested may be of nips origin. in addition the inadvertent statement that the jae are nat yee tilts es (l.c. p. 19), is a combination of C. plicata, [. genuina, Miill.-arg. (1866) and y. prostrata, Miill.-arg. (1866), again with all the defects of the ‘ gp starhon os account and most of those in the account of Boissier, ee m has been adopted the suggestion of merging three distinct species in one. statement that this ‘ epee is ae tinctorial is partly true since the two Indian species, C. prostrata, Dalz. (1861) and C. parvi- folia, Klotzsch (1862), are not tinctorial. This statement, how- ever, is not applicable to the original C. plicata, A. Juss. (1826). 3. C. Brocchiana (1.c. p. 20) is taken up from the * Prodromus’ with no change beyond that of suppressing Schweinfurth’s variety Hartmannii. 4. C. Bi staat a - ce. p. 20) is taken up about ae from the ‘ Prodro 5. C. tinctoria + va is C. tinctoria y. genuina, Miill.-a (1866). The belief as “ie the poisonous qualities of this plant is derived from Kobert (Lehrb. Intox. ii. p. 653) whose state- 88 ment, published in 1906, is due td a misreading of the original record of an incident which occurred in Persia (Kew Bulletin, 1889 p. 279), and to Kobert having overlooked the sequel to that record (Kew Bulletin, 1896, p. 233). The specimens sent from Persia as ‘ Tatuleh ’ did in reality belong to C. tinctoria, Stewart (1869) non Linn., which is identical with C. hierosolymitana, Spr. (1826), but it eventually became known that it probably was the true ‘Tatuleh ’, Datura Tatula, Willd., specimens of which were not sent, which actually caused the six deaths. ba. C. subplicata (l.c. p. 24) is C. tinctoria 6. subplicata, Mill.- arg. (1866), advanced in status. The suggestion as to the parent- age of this supposed hybrid is taken up from Boissier, not from Schweinfurth, who notwithstanding the fact that C. subplicata has lepidote capsules, has treated it, in the manuscript note to whic offmann refer, as a hybrid between Croton plicatum, Vahl, and Croton obliquifolium, Vis. (1836), both of which have stellate-pubescent capsules and are, as Schweinfurth subsequently admitted, only varities of a single species. The reference of specimens of the erect C. oblongifolia from Kosseir to the prostrate C. plicata, is copied from Boissier’s account of 1879. The erect Afghan plant, also referred in the ‘ Pflanzen- reich’ to the prostrate C. plicata, is C. hierosolymitana, Spr. (1826), with a larger number of stamens than usual.’ 6 C. glabrata (l.c. p. 24) is, as Pax and Hoffman have stated, the most easily recognised of all the forms in the genus Chrozo- phora. It is not, however, the most distinct. In their remarks on the phylogeny of this form Pax and Hoffmann have regarded it as a derivative of (. tinctoria (l.c. p. 18) but in their notes under the description of the plant they have said that it is ex- tremely distinct from C. tinctoria and is more closely related to C. verbascifolia, though still distinguishable at a glance from the latter species. Of these two views the former appears to have been adopted from Heldreich who, when he first obtained specimens, named it C. tinctoria, var. glabrata (Parnassos p. 277). The second view is undoubtedly the more satisfactory. Yet, notwith- standing the very different facies of the plant, due to the almost complete absence of the characteristic loose woolly tomentum, a careful examination of the original type leads to the impression that this plant is no more than a nearly glabrous sport or condi- tion of C. obliqua, A. Juss., hardly deserving to rank as a dis- tinct variety. Y oblongfolia (l.c ee Del. with the erroneous identification suggested by Miiller in this species has been perpetrated in the ‘Pflanzenreich ’ and a new inadvertence has been introduced by the reference to C. ob- aaa Mae See 89 longfolia of a specimen of C. Hern en dgonan ap collected by Schlagintweit at Rawalpindi in the Punj C. verbascifolia (l.c. p. 26) is taken up from Boissier’s account of 1879, and thus includes C. tinctoria a. verbascifolia, Sea 5 (1866) and p. fie Boe ie Miill.-arg. (1866) two or aillon % But the tendency to confuse C. hierosolymitana with C. verbasci- folia has hardly been greater than the complementary tendenc to confuse C. verbascifolia with C. tinctoria, of which the ra ture of the genus offers no example. The dubiety suggested as to the identity of Croton obliquum, Vahl (1790) does not really exist, for the type of Vahl’s species is also the type of Croton argenteum, Forsk. (1775) non Linn., and Croton argenteum, Forsk., as the specimen in the Copenhagen herbarium shows, is the species eaiee Willdenow in 1805 described as Croton verbascifoliu 9. C. sabulosa cae p- 27) is taken up from the ‘ Prodromus.’ SYSTEMATIC SYNOPSIS. may, it is ig vite serve the double purpose of facilitating com-. parison with previous accounts and of a a hn ae geographi- cal distribution of the various ieipleebte form CHROZOPHORA Neck. § I. Tricnocarpa. Capsula pilis stellatis induta, nunquam lepidota. od. ae Paz et K. Hoffm. in Engl. Pflanzenr. IV. 147,- 70 ee (1912). Antherae 3-verticillatae cane pars ibera filamentorum oes nie ste” te phora, § 1, Miill.-arg. in DC. Prodr. xv. 2. p. 747 (1866). * Six species have from time to time been recognised among the members of this group :—1. Chrozophora plicata, A. Juss. (1826), which is Croton plicatum, Vahl (1790), based on Croton tine- torium, Forsk. Cent. vi. 1775) non Linn.; 2. Chrozophora Rottleri, A. Juss. (1826), which is Croton Rottleri, Geis. (1807) ; 3. ae Burmanni, Spr. (1826), based on emo hastatum rm. 1768 n PUI (1805) nee Vahl. We now know, however that 2. C. Rottleri and 3. C. Burmanni are only conditions of the same species, and that 4%; Voleghafolie is only a variety of 1. C. plicata. 90 CLAVIS SPECIERUM INTER ‘ PLICATAS.’ Capsulae cinereo-nigrescentes ; stigmata aap isagtd petala lutea; caules prostrati : Folia basi eaactiaoas: pilis stellatis aepe stipitatis induta ; herbacea.. 1. C. prostrata. Folia basi 2-glandulosa, pilis stellatis sessilibus induta; suffruticosa ... 2. C. parvifolia. Capsulae purpurascentes; stigm ata rubra; folia basi 2-glandulosa — Racemi quam folia eis proxima breviores; capsulae violaceo- purpureae ; petala ey: — Caules prostrati :— ear agereat pedunculi brevis- mi, basi saepe foliati ... 3. C. plicata. Polis pokey pedunculi dis- tincti, nudi ... eee iy pt » var. obliquifolia. Caules erecti an Pi Fah es var. erecta. Racemi quam folia eis proxima longiores; capsulae rubro-pur- pureae ; petala sordide lutea; folia saepe majora ; caules erecti . 4. C. Rotileri. 1. Chrozophora prostrata, Malz. in Dalz. et Gibs. Bomb. Fl. p- 233 (1861). Annua, prostrata; folia radicalia rosulata, subper- sistentia, basi eglandu osa, pilis stellatis magis minusve stipi- tatis obtecta; petala lutea; stigmata aurantiaca; capsula matura cinereo-nigrescens, haud tinctoria.—Cooke Flor. Pres. Bomb. 1. p. 607 Aare ee Afr. cit. exclud. Croton plicatum, Ham. ex Wall. Cat. Lith. n. 7716 C et n. 7716 D ane ae nec Vahl. Croton Seca Wall. Cat. Lith. T716 partim (1830) nec Linn. af aad lanuginosum, "Behan m. ex Sch . Pl. quaed. Nilot. 0 (1862) nec Baill. Chrozo- re plicata [3. genuina, Mill, ave in Mtoe Bigot Ve p. 747 (1866) quoad spec. Ind. cit. sed syn n. exclud. C. plicata y. prostrata, Miill.-arg. 1.c. (1866) aumad syn. Dalz. et Klotzsch tantum. C. plicata, 3, Hook f. Fl. Brit. Ind. v. p. 440 (1887). C. plicata a, Watt, Dict. Econ. Prod. Ind. 260 (1889). C. plicata Stewart, Panjab. Pl. er parti } a toe ae King, Pl. N.W. Prov. p. 15, sabe (1870); Pax et K offm Engl. Pflanzenr. IV. 147, vi. p. 19, nda (1912) nec A. 5 ss. This species is widely distributed in Peninsular India from the Panjab and the Upper Gangetic Plain southward to Coro- mandel. In the north-east it does not extend to the Lower Gan- getic Plain, in the south-west it does not extend to Malabar. It is a small procumbent annual found in damp situations such as » eT leaves dictinguiah it at once from th which it has hitherto been customary roi associate it. In the 91 African plants in question the leaves are 2-glandular at the base, the capsules are violet-purple and yield a dye, the stigmas are red and the petals, so far as is known, are always pink. Norruern Inpra. Panjab: Jalapur, Vicary (Croton stelli- gerum, Vicary MSS.)! Lahore, 7. Thomson! Multan; Royle! Edgeworth, 8028! Upper Gangetic Plain: Gorakhpur, Duthie, 2249a! Behar; Patna, Hamilton, 2107 (Wall. Cat. n. 7716 © partim)! Naoranga, near the River Son, Jacquemont, 112! 200! Western Inpra. Gujarat: Rajkot, Birdwood! Concan: Northern district, Ritchie! Deccan: Aurangabad, Ralph, 22! Egadon and Vouanagram, Campbell! Kolapur, Ritchie, 1345! North Canara; Bomanhulli, Talbot, 270! 359! without precise locality, Stocks! Centra Inpia. Malwa: Guna, King, 34! without precise locality, Jerdon! Central Provinces: Nimar; Khandwa, Duthie, 8421! without precise locality, R. Thompson, 230! Sourwern Inpta. Mysore: near Mysore, Heyne (Herb. Rottler, also Wall. Cat. n. 7716 B partim)! Carnatic: Pierwandi, Wight, 63 (Wall. Cat. n. 7716 D partim)! Kistna; Kaduva Kad- hura, Gamble, 12562! Chingalpat; Vellapura Choultri, Griffith! adras, G. Thomson, 110! Salem; Shevaroi Hills, Perrottet, pk Pondicheri, Lecomte! Raymond! Tranquebar, Koenig! ottler! 2. Chrozophora parvifolia, Klotzsch ex Schweinf. Pl. quaed. Nilot. p. 10 (1862). Perennis, prostrata; folia radicalia_mox evanida basi insigniter 2-glandulosa, pilis stellatis sessilibus parce obtecta; petala lutea; stigmata aurantica ; capsula matura cinereo-nigrescens, haud tinctoria.—Croton plicatum, Willd. Sp. Pl. iy. 1, p. 538 (1805), quoad spec. Ind. cit. sed syn. omn. ex- clud.; Klein ex Schweinf. Pl. quaed. Nilot. p. 10 (1862): nec Vahl. Chrozophora plicata y. prostrata, Mill.-arg. in DC. Prodr. xv. 2, p. 747 partim (1866) quoad syn. Klein et Klotzsch tantum. Chrozophora plicata 2, Hook. f. Fl. Brit. Ind. v. p. 410 (1887). _ This species, like the preceding, to which it is nearly allied, is confined to Peninsular India, where it is widely spread from Scinde, the Panjab, and the Upper Gangetic Plain southward to Coromandel, but while present in Scinde, whence C . prostrata has not so far been reported, it has never been gathered in the Deccan or in any part of Central India except Bundelkhand. A procum- bent plant like the preceding species, it has in almost every 1n- stance been gathered as a woody plant of more than one season's In thei lent monograph of the genu Chrozophora, Pax n their excellent monograp ge re cae possibility of the occurrence within it of natural hybrids. The might be regarded as a natural cross between C. Rottleri and C. prostrata, in which the glands at the base of leaf beneath and the 92 does.so in Northern India, the latter species seems to do so in Southern India. The chi ef difficulty connected with the accept- ance of the view lies in the fact that C. parvifolia occurs in Scinde, where C. prostrata has not yet been found. NortHern Inpia. Scinde: without eee locality, Stocks 547! Punjab: Lahore, 7. Thomson, 1508! Ambala, Edgeworth, 126! Sirsa, Drcnmond, 2352! near Tusham Rock, Herb. Drummond, 3060! 3062! ’Khanak Hill, 1000 ft., Herb. Drum- mond, 3061! Hissar, Ram Baksh in ‘Herb. Drummond, 6346 ! Gurgaon, Ram Baksh in Herb. Drummond, 6345! Karnal, Drummond, 6348! Upper Gangetic Plain: Oudh; Kheri, PANS 22494 ! Lucknow, Anderson! Farukhabad; Fatehgarh, Griffith ! —— near Mirzapur, Griffith! without precise locality, Royle! Cenrrat Inpra. Bundelkhand: without precise locality, Hdge- worth, 8029! SourHEeRN Inp1a. Madras Presidency: Anantapur; Bukkapa- tam, 1500 ft., Gamble, 21167! Chingalpat; Tiruvalur, Klein 396 in Herb. Willdenow! near Madras, Gwifith! Salem ; Shevaroi ne Perrottet, 77! n the Edinburgh Herbarium there is an Indian specimen of gioco parvifolia, collected by Ritchie, which has not been localised. The question as to the possible provenance of this speci- men (Zitchie n. 670) will be discussed along with the similar problem involved in the case of another (Ritchie n. 671) which belongs to Chrozophora hierosolymitana, and is also unlocalised. It may be stated here, however, that probably neither specimen came from any part of the Bombay Presidency, which was the scene of Ritchie's activities as a collector. The locality Malacca, cited by Miiller, is the result of some misapprehension; the epee which was collected by Klein, came from Southern India 3. Chrozophora plicata, A. Juss. ex Spr. Syst. Veg. ili. p. 850 (1826) syn. Lamk exclud. Annua, saepissime prostrata raro in Africa orientali erecta; folia radicalia mox evanida, caulina basi 2-glandulosa, pilis stellatis magis minusve stipitatis obtecta; petala pallide punicea; stigmata rubra; vised gr matura violaceo- purpurea, distincte tiactoria; racemi congesti quam folia summa breviores. —Prain i in — Fi. Trop. Afr. vi. 1. i Pe 834 (1912). 93 catchment-areas of the Senegal River, the Niger -Benue basin, the dvainage area of Lake Chad, the Nile basin, and again in the catchment-areas of the Zambesi and the Lim mpopo. “It has not yet been reported trom any ages in the Congo basin or in that of the Cunene or the Orange Riv Three very easily age varieties may be recognised. a. typica, Prain, (1912). Prostrata; folia caulina saepis- slime longiora quam fie margine undulata ; pedunculi perbreves saeplus versus basin foliiferi.—Croton tinctorium, Forsk. FI. Aegypt.-Arab. Cat. Aegypt. n. 490, p. lxxv. et (dubitanter) Cent. vi. p. 162 (1775), nec Linn. Croton plicatum, Vahl, Symb. Bot. p-. 78 (1790); Willd. Sp. Pl. iv. 1, p. 538 partim (1805) quoad syn. Vahl tantum; Geis. Crot. Monogr. (plicatus) p. 70 (1807), syn. Lamk et Burm. f. omnino syn. Vahl pro parte maxima excl. ; A. Juss. Tent. Gen. Euph. p. 28 (1824). Chrozophora plicata, A. Rich. Tent. Fl. Abyssin. 11. p. 252 (1851) excl. syn. Burm. f. ; p. 67, partim (1860) ; Boiss. Fl. Orient. iv. p. 1140 partim (1879) Engl. Hochgebirgesfl. Trop. Afr. p. 283 partim Sue a: in Istit. Bot. Roma, vi. p. 183 (1895); Broun, Cat. es p. 72 (1906); Pax et K. Hoffm. in En gl. Pflanzenr. V. 147. 5 p. 19 partim (1912). C. obliqua, Schweinf. Pl. quaed. Nilot. . 10, t. 3 (1862), et in Beitr. Fl, Aethio p p- 99 (1867), non A. Juss. i noe 3 c in Trans. Linn. Soc. xxix. p. 146 (1875); Aschers. et Suh eelat. Ill. Fl. Egypt. : a (1887); Schweinf. in Bull. Herb. Boiss. vii. App. 2, p. 806 (1899). C. prostrata, Schweinf. Beitr. Fl. Aethiop. pp. 235, P9362 (1867), non Dalz. Seneca, River Basty. Senegal: Dagana, Lepriewr! Podar, Mathieu! Nicer-Benve Basry. Cameroon: Garua, on sandbanks in the bed of the River Benue, Ledermann, 3239! Laxe Cuap Basty. Northern Nigeria: Bornou; Kuka, near Lake Chad, Vogel, 32! French Congo: Kusseri, on the River Shari, Foureau, 3006! Nite-tanp. Sudan: banks of the Nile; at about 16° N. Lat., Speke & Grant! Nuer; on the White Nile, Brownall! at Geitena, Schweinfurth, 898! Sennar: Abu Sugra, Schweinfurth | Wolet Medine, Kotschy, 473 partly! Kordofan, Colston, 52! 63! Salati, lue Nile, Muriel, S/37! Abyssinia : Goelleb, ' 1694! Amhara; Jenda, Steudner, 533! by the Tacazza River, Petit! Quartin-Dillon § Petit, 213! Weg Sch not seen}. Nu Ri eer Tratan tapes : arr Courbon, 438! nbn orn- miller, 10949! Lower Egypt: Giz orskal! n ai izeh, Ff ro, ! near Ehrenberg! Schweinfurth, 137! Wichura, 3075 ! Pfund! ne Shales Alley; SScanhennt between Cairo and Roman toh, 94 Tomard! Zagazig, Ball! Benha, Schweinfurth, 837! Damietta, ratatlnderee Se Samit Naim, Samaritani, 3612! Torrah, Kotschy! Boissier! without precise locality, Delile! ahem Montbret! Zambrst River Basin. Rhodesia: andbanks in rivers, Allen, = on the banks of the Santee Wilde in Herb. Transvaal, 9054 Asia. Syria ss Piahatbiia: Jerusalem, Meyers, 95 partly ! with- out precise locality, Olivier § Bruguiére! (3. obliquifolia, Prain, l.c. p. 835. Prostrata; folia caulina a onga ac lata, margine repanda; ons ae distincti, nudi. ‘roton plicatum, Sieb. Avis, herb. Aegypt. p 7; advers. p. 8 (1821), vix Vahl. Croton obliquifolium, Vis Pl. quaed. Aegypt. ac Nub. p. 39, t. 7, fig. 1 (1886). Chrozophora obliquifolia, Baill. Etud. gén. Bopherd. > 322 (1858). C. plicata, Baill. in Adan- sonia i. p. 67, partim (1360) ; Schweinf. Pl. quaed. Nilot. p. 11, t. 4 (1862), syn. Vahl. excl.; Klotsch in Peters, Reis. Mossamb. Bot. ii. p. 576 (1863) ; Schweinf. Beitr. FI. Aethiop. pp. 36, 262 (1867); Boiss. Fl. Orient. iv. p. 1140, partim (1879) ; Aschers. et Schweinf. Ill. Fl. Egypt. p 168 (1887) ; Engl. Hochgebirgestl. Trop. Afr. p. 283, partim mm. (1892) ; Pax in Engl. Pfl. Ost-Afr. C. p. 237 (1895); vix A. gets = ie tinctoria, Klotzsch, l.c. p. 99 (: pula. Rick Basin. Benipal’ Dagar, Phiri side |: Lake Cuap Basrn. Northern Nigeria: Kuka, near Lake Chad, V a 3! Nine-ranp. Sudan: Burri, Schweinfurth, 832! near Shendy, Sere iarih. 737! North Ipsambul, Scott Elliot, 3418! Sennar: Fazokl, Ehrenberg! near Abu Harrah, Schweinfurth, 833! 834! Wolet Medine, Kotschy, 473 partly ! White Nile, d Arnaud ! Korotan, Pfund! Daliton, 14! Abyssinia: Bege- meder; by River Reb. Schimper, 1355! Eritrea, Abai, Stecker! "acainfelaen Rohlfs & Stecker! Nubia: Dongola, . ge and Sabou, Kralik! without ore mre | Pfund! Upper : Assiout, Sieber! Lusson, 309! Bromfield ! El Haameh, oh O16! Middle Nile, Wenne! Deiiders: Hart- mann ! Assouan, Kiigler! Farshut, Schweinfurth, 854! Fie Wilkinson! near Gournass, Letournour, 302! Lower Egy Pyramids, Sieber! Pilgrims’ Lake and Birguet-el-Agio, Ridked | Cairo, Heldreich! Wiest! ten miles south of Benha, Schwein- furth, 856! Wadi Dugla, beween Cairo and Suez, Schweinfurth! Suez, in clover fields, Rensch in Herb. Hi ‘Ladbewidt 89! Torrah, Kotschy! without precise locality, Olivier §& Bruguiére! Auche er, 2005! Galopin! Boissier! Wiest in Herb. Schimper, 517! Fischer! ZaAMBESI River Basry. a penrebi que: Sena, Peters, 8! Gonon- goza; Sungine, Vasse, 403! y. erecta, Prain. Erecta; folia caulina aeque longa ac lata, margine undulata; pedunculi perbreves, nudi. Lisporo River Bast. a Sareea Mazambo; eastern bank of the River Limpopo, Bevje 95 The two first ential are prostrate branching herbs with a stout root. In the Sudan a blue dye is obtained from the capsules. The stems yield a strong fibre which is not easily separable (Broun). Both the seeds and the leaves have purgative properties. All three varieties occur on river banks and on sandbanks in river beds. The two prostrate varieties occur also in depressions, more especially in cotton soil; the second variety is, besides, a common weed of cultivated ground and palm groves The specimens of the typical plant collected by Clive and Bruguiére which have been localised as from Syria have no original field-note. The same is the case with the specimens of p. ee re obtained by the same collectors, which have been localised as from Egypt. It is therefore permissible to conjecture that some error has taken place with regard to the Syrian habitat. This conjecture is not, however, necessarily correct because it has recently been ascertained that typical C. plicata does occur now, near Jerusalem, perhaps as a recent introduction. In the Limpopo lecality, which is close to the Tropic of Capricorn, the species is plentiful, but only the third variety, Dr. Beijer informs us, is to be met with there. The statement relied upon in 1912 that the petalsin $8. obliquifolia may be yellow (Fl. Trop. Afr. vi. 1. p. 835) appears to i incorrect; they are pink, as in a. typrca and in y. erecta. The statement that the plant is not tinctorial (Pflanzenr. IV. 147. vi. p. 18), though correct as regards the two Indian species there merged in C. plicata, does not apply to any of the forms of C. plicata itself. 4. Chrozophora Rottleri, A. Juss. ex Spr. Syst. Veg. iil. p. Sti (1826). agp vel raro diutina, erecta; folia omnia Satins, pliis minusve 3- iia onnunquam undulato-crenata, rarius ovata subacuta vel acuta pprcehs subintegra, oe stellatis sek Ribs aspera; petala lutea; stigmata rubra; capsula matura rubro- purpurea, valde tinctoria ; racemi pro. genere dete I. 19 (1912). Ricinoides malabarica surattensis, Garcin ex Burm f. Fl. Ind. p. 305 [205] Sats Croton tinctorium, Burm. f. l.c. p. 304 [264], t. oF fee (1768); Roxb. ex Wall. Cat Lith, n. TOG ASS apres I (18 830), nec Linn. Croton hastatum j3., Burm. f. | 5 [205], t. 63, re 1 (1768), non C. hastatum, Linn. Croton ieee B. ety., Lamk Eneye. Meth, ii. p. 214 (1786). Croton moluccanum, Willd. Sp. Pl. iv. 1, p. 551 (120); non Linn. Croton Rottleri, ee Crot. Monogr. D. 54 (1807) ; - Juss. Tent. Gen. Euph. p. 28 (1824). Croton asperum, Koen. e Roxb. Hort. eg p- 104 (1814); Wall. Cat. Lith, n. 7716 C Le. : 826). Chrozophora plicata, Voigt, Hort. Bulan. SE, ab (1846); Dalz. et Gibs. Fl. Bomb. p. 233 (1861); 96 ( Ind. ii. p. 620 (1889). Yournesolia plicata, O. Kuntze, Rev. Gen. Pl. ii. p. 621 (1891). A distinctively Indian species, very rare in the Punjab, where it is replaced ‘by the non-tinctorial C. parvifolia, Wlotsch, wards to Coromandel and Northern Ceylon; absent from Malabar and Southern Ceylon. From the Upper Gangetic Plain it extends to the Lower Gangetic Plain, thence into the valleys of the Brahmaputra and the Surma and again thence into those of the Irrawadi in Burma, and of the Meinam in Siam. Usually an annual-field weed throughout its area and always so in Northern India and Western Indo-China, Chrozophora Rottlert sometimes lasts for more than one season in dry waste non Vahl; Croton asperum, Koen.}; the form with narrower sub- entire somewhat acute leaves is Chrozophora tinctoria, Cooke, non A. Juss. (Croton tinctorium, Burm. f. non Linn.); the perennial shrubby condition with rather smaller, rounded but not very open. Ceyton. Trincomali, Glenie in Herb. Thwaites, 3854! Sournern Inpra. Carnatic: Tanjore; Tranquebar, Koenig! Heyne in Herb. Wallich! Rottler or Klein in Herb Wallich! Coimbatore, Beddome, 7293! Anamalai, Wight, 2313! 2613! eed ie near Devanur, Klein! Haskancharni, G. homson, 24! Madras, Shuter! Griffith! Wight, 62! Cuddapah ; Poramanilla, 500 ft., Gamble, 11097! Mysore: Salikraman, G. Thomson, 428! Circars: Samalcotta, Rorburgh! without precise locality, Russell in Herb. Wallich! Beddome, 7294! Western Inpia.. Deccan: Dharwar, Cooke! Vouanagram, 97. Campbell! Aurangabad, Ralph, 21! Poona,. Stocks! Cooke! Woodrow! Concan: North Concan, Ritchie! Surat, Garcin (Ic. Burm.)! Gujarat: near Gujarat, Woodrow! Scinde: near Karachi, Dalzell! between Karachi and Tatta, Schlagintweit, 10995! Centrat Inpra. Malwa: without precise locality, Jerdon! Central Provinces: Nimar; Khandwa, Duthie, 8422! without pre- cise locality, R. Thompson! Nortuern Inpra. Panjab: Lahore, 7. Thomson! Upper Gangetic Plain: near Saharanpur, Royle! Gonda; Barhaiwa, Inayat! Pilibhit, Duthie, 22495! Banda, Bell, 85! Mirzapur; banks of the Ganges at Shahganj, Hooker, 545! Griffith! Behar; Naoranga, Jacquemont, 12! 113! Domdoha, Hamilton, 2106! Lower Gangetic Plain: Bengal; Maldah, Vicary! Kushtia, Kurz! Manbhum, Campbell in Herb. Wait, 9811! Sonthal Par- ganahs, Campbell! Caragola Ghat, Kurz! Chandernagore, Herb. Calcutta! Serampore, Carey! Sibpur, Roxburgh! Wal- lich! Gaudichaud! Griffith! T. Anderson! Kurz! 24-Parganahs: Takeo, Clarke, 34845! Sundribuns, Clarke, 33360! Jornadun! Inpo-Cu1na. Assam: Surma Valley, Hooker, 301! Golaghat, Jenkins! Simons! Burma: Kyouk-zeik, Buchanan! Pegu, Roz- burgh! Paghanmyo, Wallich. 7716 I! Myingyan, Griffith! J. Anderson! Prome, McClelland! Myanoung, Gamble, 2828! Tonkyeghat, Aurz, 1545! Sagain, King’s npr OL em 5H King’s Collector! Siam: Radboeri, Tezjsmann, depends on an inadvertence; the specimen cited came from Southern India, where it was collected by Klein. € 2. Graciles. Antherae 2-verticillatae, quam pars libera fila- mentorum breviores.—Chrozophora, §2, a., Miill.-arg. in DC. Prodr. xv. 2 48 (1866). At one time two species were recognised in this group:—l. Chrozophora gracilis, Fisch. et Mey. (1839) and 2. Chrozophora in 1866 used the name recognised by Bunge; also by Boissier, who in 1879 used that published without a description in ‘ D 98 Regel and Herder in 1869 followed the recently published work of Miller; Kuntze on the other hand in 1891, when changing the generic name to Tournesolia, followed Boissier as regards the specific one. In 1912 Pax and Hoffman have once more adopted for the only species of the group the name Chrozophora sabulosa which, though not the oldest, is the one under which a descrip- tion was first given. 5. Chrozophora sabulosa, Kar. et Kir. Bull. Soc. Nat. Mosc. xv. p. 446 (1842). Annua, erecta; folia basi eglandulosa, pilis stellatis sessilibus griseo-tomentella ; petala lutea; stigmata rubra; capsula matura rubro-purpurea, tinctoria; semina laevia. —tTLedeb. Fl. Ross. iii. 2, p. 581 (1850); Bunge, Rel. Lehm. p. 314 (1851); id. in Mém. Acad, Pétersb. vii, p. 490 (1854) ; Baill. Etud. gén. Euphorb. p. 322 (1858); Mill.-arg. in DC, Prodr xv. 2, p. 748 (1866); Regel et Herd. Enum. Pl, Semenov. pars 4, p. 94 (1869); Pax et K. Hoffm. in Engl. Pflanzenr. LY. 147, vi. p. 27 (1912). Chrozophora gracilis, Fisch. et Mey. Bull. Soc. Nat. Mose. xii. (Kar. Enum. Ture.] p. 171, nomen (1839); Fl. Orient. iv. p. 1140 (1879). Tour- CentraL Asta. Soongaria: between Sassky Pastau and the Arganiti Mountains, Karelin § Kirilow, 1941! on the River Tlu, Schrenk, 24! on the River Ili and in the Ili plains, Seme- now! Turkestan: Kisil-kum; between the River Kuwan and the River Jan-darya, Lehmann, 1249! Suidun, A. Regel! Askabad, Litwinow, 172! Sintenis, 416! without precise locality, Turczan- inow ! § II. Leprpocarra. Capsula_ lepidota nec stellato-pilosa; antherae semper 2-verticillatae quam pars libera filamen- torum parum longiores.—Chrozophora, § 2, 8, Mill.-arg- in DC. Prodr. xv. 2, p. 749 (1866). € 3. Tinctoriae, Pax et K. Hoffm. in Engl. Pflanzenr. IV. 147, vi. p. 21 (€. sabulosa ewel.), pro sectione (1912). Petala sordide lutea; stigmata rubra; capsula matura purpurea, tinctoria, squamis discretis margine denticulatis obsita; semina aspera.—Chrozophora, § 2, [3., 1., Mill.- arg. l.c. (1866). The number of forms admitted in the group Tinctoriae has varied somewhat. Tournefort in 1703 recognised two species of Ricinoides ex qua paratur Tournesol gallorum, which he dis- tinguished as folio oblongo et villoso, and as folio serrato non villoso respectively. Willdenow in 1805 nominally distinguished three :—Croton tinctorium, Linn. (1753); C. obliquum, Vahl (1790); and his own C. verbascifolium, based on Tournefort’s Ricinoides . . . . folio oblongo et villoso. In reality, however, Willdenow only dealt with two species because CU. verbascifolium is the same as the older C. obliquum. Geiseler LS eee 99 Since the resuscitation of Chrozophora it has been usual to enumerate five forms with characters that mark them as members of this group. In 1826 Sprengel cited the four mentioned by A. Jussieu as Chrozophora tinctoria, i.e. Croton tinctorium, Linn. (1753) ; obliqua, i.e. Croton obliquum, Vahl (1790) ; verbas- ctfolia, i.e. Croton verbascifolium, Willd. (1805); and oblongi- folia, i.e. Croton oblongifolium, Del. (1812); but added a fifth, booth hierosolymitana, i.e. Croton oblongifolium, Sieb. , nou Del. In reality, therefore, Sprengel dealt with only four dis- tinct members of the group Tinctoriae. e five enumerated by Baillon in 1858 were :—Chrozophora integrifolia, Bunge (1851); tinctoria, i.e. Croton tinctorium; obliqua, i.e. Croton obliquum; verbascifolia, i.e. Croton verbascifolium; and oblongifolia, i.e. Croton oblongifolium. Though the list of Baillon agrees as regards number of species with that provided by Sprengel, the species are not the same. Baillon has not accounted for Chrozo- phora hierosolymitana, and as C. integrifolia is merely a new name for the species to which the two names obliqua and verbasci- folia alike belong, we have in his list only three species of the group Tinctoriae. In 1866 Miiller only recognised two species, Chrozophora tinc- ua. As, however, he has, under C. tinctoria, Miill.-arg., discriminated four varieties, viz. :—verbascifolia are legitimate and valid; it has the disadvantage as compared with theirs that C. obliqua mitana; his last form and third species, C. obliqua, is not the plant so named by Jussieu, but that so named by Miiller. In : p2 100 1891 O. Kuntze followed Boissier in recognising three species as Tournesolia obliqua, tinctoria and verbascifolia, the last as doubtfully a distinct species. In 1912 Pax and Hoffmann reverted to the number of forms recognised by Sprengel, Baillon and Miiller, the five enumerated by them being (1) Chrozophora tinctoria, A. Juss. (Croton tinc- torium, Linn.) ; (2) C. subplicata Pax et K. Hoffm. (C. tinctoria 6. subplicata, Mill.-arg.), which they follow Boissier in regarding as a hybrid; (3) C. verbascifolia, A. Juss. (Croton verbascifolium, Willd.) which they consider, as Jussieu did, to be identical with Croton obliquum, Vahl; C. glabrata, Pax et K. Hoffm., based on C. tinctoria, var. glabrata, Heldr. (1899); and C. oblongifolia, A. Juss. (Croton oblongifolium, Del). This treat- ment, like that of Miiller, has the advantage that all the forms recognised are valid though, perhaps, not all of them are distinet species. It has, however, the disadvantage, as compared with the arrangement of Miiller, of following Boissier in merging C. hierosolymitana in C. verbascifolia. It is true that C. hierosoly- mitana has flowers like those of C. verbascifolia. But it is not the case that it is difficult to distinguish between these two forms; the difficulty is rather to distinguish, without careful dissection, between C. hierosolymitana and C. tinctoria. When we add C. hierosolymitana, Spr. to the forms recognised by Pax and Hoffmann we find therefore that in the group 7inctoriae we have six forms, belonging to four species, as indicated in the subjoined conspectus. .- CILAVIS SPECIERUM INTER “ T'INCTORIAS.” Capsulae maturae distincte muricatae :— Folia quam lata duplo longiora; capsulae maturae violaceo-pur- pureae ; frutex erectus ... ... 6. CO. oblongifolia. Folia quam lata minopere longiora ; capsulae maturae rubro-pur- pureae; herbae :— Antherae 9-11, saepissime 10; folia adulta saepius scabrida, raro molliter tomentosa :— Caulis erectus... wile ... 7. C. tinctoria. Caulis prostratus... a Say 2! » var. subplicata. Antherae 5-6, raro 7-8; folia adulta velutina rarius parce . tomentosa... _ ... 8. OC. hierosolymitana. Capsulae maturae vix vel minopere muri- catae, rubro-purpureae; antherae 4-5, raro 6-7; folia quam lata dimidio longiora :— Folia utrinque laxe molliterlanuginosa 9. C. obliqua. Folia utrinque glabrescens vel glabra.. 9b. var. glabra. 6. Chrozophora oblongifolia, A. Juss. ex Spr. Syst. Veg. iii. is 850 (1826). Hornotina herbacea, diutina fruticosa, erecta ; lia. subffoccoso-tomentosa quam lata duplo longiora, margine 101 saepissime inciso-lobata vel lobata; ppaheres saepius 4-7; capsula ruatura _distincte muricata, coeruleo-pu urea. Kitud. gén. Euphorb. p. = (1858); T. And. in Journ. Linn. Soc. v. Suppl. [ Flor. oki p- 36 (1860) ; Pax et K. Hofim. in Engl. Pflanzenr. IV. 147. p- 25, loc. Panjab et syn. Forsk. ac Wall. necnon syn. C. obliga, Juss. excel. (1912); Prain in Dyer Fl. Trop. Afr. vi. 1, p. 836 (1912). Croton tinctorium? Forsk. Fl. Aegypt.-arab. n. 563 Cat. Arab.-Yem. p. exxi. (1775), nec Linn. eae oblongifolium, Del. Descr. Egypt. Hist. Nat. . Egypt. p. 139, t. 51, fig. 1 (1812); A. Juss. ey Gen. Haph, p- 8 ae 24). Chrozophora obliqua, Miill.-arg. in DC. Prodr. xv. 2, p. 749, syn. Wall. 7716 G excel. (1866); "Solwainl: Beitr. FI. Aethiop. p- 235, Aufz. p. 262 (1867); Boiss. Fl. Orient. iv. p. we in Schweinf. excl. (1879); Hook. f. Fl. (1887); Balf. f [ Bot. Socotr ra | Trans Roy. Soc. “Edin. 3 XXxl. p- 277, var. frutescens, Schweinf. ee (1888); Penzig in Atti Congr. Bot. Genova, p. 359 (1 ; Pax in Ann. Istit. Bot Bae vi P 183 (1895); Schweinf. in Bull. Herb. Boiss. vii A 06, var. st tifo ia ac var. incisa incl ete p- 607 (1906) ; Blatter ns Rec. Boi t. Sui rv. Ind. vii. _ [Flor. Ades n| p. 332, syn. Wall. 7716 G excl. (1915); non A. Piet nec Schweinf. (1862). bidup hors Te Ce Schwein. Beitr. Fl. Aethiop. p. 36 (1867); Balf. f. lc. (1888); A. Juss. Chrozophora tinetoria, var. wabplicn ata, “Boiss. os tae et quoad loc. ha ee oe er a Ph tantum (1879); Terraciano in Ann. Istit. Bot. Roma, v. p. 98 Abs: es nec Mill.-arg. Chrozo- phora subplicata, Pax et K. Hoffm. l.c 24, partim et quoad loc. Kosseir [Schweinfurth] ‘ait M 1912). Tournesolia obliqua, Franch. in Morot, pete Bot. i. p. 135 (1887); O. Kuntze, Rev. Gen. ii. p. 621 (18 This species is chatactanttie of, eas is almost herpes con- fined to the sublittoral zone of the north-east c and the south-west coast of Asia, along both sic of the Red Sea and the Gulf of Aden. On the African shore it extends from Suez to Obok and Jibuti, thence to Ras Asir and Socotra. On the Asiatic coast it extends from the Sinai ha meso to Perim, thence to Muscat. Outside the area it has been once met with on the coast of Scinde but in no intervening omit; it may therefore in India be only a species introduced from Aden or Muscat. In the Sinai “Parken the ash of the fruit is used in the treatment of suppurating wounds. AF . Egypt: Ajeraud, Delile! between Ajeraud and Suez, Letournoux! Suez, Schweinfurth Trigari, Parlatore! Wadi Nachel and Hendossa, near Kosseir, ‘Klunzinger! Wadi om Schweinfurth, 945! 952! without precise locality, ? Lippi Herb. Jussieu, 16255! Nubia: headland of Jebel ‘errajeh, near Berenice, Schweinfurth, 943! Jebel Garab and Jebel Dyb, 102 Lusson, 493! coast at Mirza Elei, Schweinfurth, 936! at Mirza Abu Amaneh near Ras Ranai, Schweinfurth, 937! Bent! Shadeh, rear Suakin, Schweinfurth, 7 bis! Eritrea: Massowah, Deflers! Samhar, Rensch in Herb. Hildebrandt, 738! Arkiko, Schweinfurth § Riva, 142! Anfilah, between Ras Madir and aressan, Terraciano. Abyssinia: Bembea, Schimper, 1692! Meda, Schimper, 1692 bis! Edd, Courbon! Ennacoullon, Cour- bon, 323! Airuri, Stecker, 29! without precise locality, Salé! Plowden! Somaliland: Obok, Faurot! Berbera and vicinity, Boivin, 1074! Révoil, 181! Drake-Brockman, 691 5821 533! 534! 535! coast near Lasgori and hills up to 3000 ft., Rensch in Herb. Hildebrandt, 860! Tokosha, Hilenbeck, 146! Socotra : near Tamarida, Schweinfurth, 358! Balfour, 133! near Galonsir, Balfour, 6441: 380! Wahab! Welsted! Wadi Tihman, Kneucker! Arabia Hedjaz; between El Wijh and Hamz-do-Kudian, Burton! Jeddah and neighbourhood, Fischer, 7! 58! Kotschy in Herb. Schimper, 993! Zohrab, 91! Rensch in Herb. Hildebrandt, 155a! Kruijt, 47! 79! Botta in Herb. Drake! Yemen; Tihama, 250 dt., Deflers! Gunfuden [Kunfuda], Ehrenberg! Hais, Botta! Bové, 237! Lohaja, Forskai! Wadi Jultl, near Juma, Lunz, 31! Kamaran Island, Faurot! Aden; Shamsen, Hooker, 100! Deflers, 62! Aden, T. Thomson! Wykeham Perry! Balfour! Schweinfurth, 29! Ralph, 668! Wiesner! Hildebrandt, 783! Kensch in Herb. Hildebrandt, 784a! Marshall Ward! Biggari Valley, Deflers! Maala plain, Lunt, 347! Shukra, Schwein- furth, 70! Oman; Muscat, Leclancher, 32! Aucher, 5296! Bornmiiller, 592! Bent, 101! Dubuc! India: Scinde; without precise locality, Stocks! The locality Mayotte in the Comoro group given by Miiller on the authority of a specimen in the Lenormand herbarium is erroneous. The field note which accompanies a specimen of this particular gathering (Boivin, 1074) shows that it was obtained by Boivin at Berbera in Somaliland. The Panjab locality cited by Pax and Hoffman and the Indian locality indicated by Miiller the citation by Pax and Hoffman of Croton argenteum, Forsk. (1775) non Linn.; Croton obliquum, Vahl (1790) and Chrozo- phora obliqua, A. Juss. (1826). The same actual specimen in the Copenhagen herbarium serves as the type both of Forskal’s species and of that of Vahl; the type of Jussieu’s species is an- other specimen of the same plant in the Jussieu herbarium writ- ten up as Croton obliquum by Vahl himself. Clearly therefore all three names must be cited under one species, not distributed 103 confusion with regard to the name which the shrubby form of mie mae should bear. The confusion between it and Chrozo- obliqua arose from some misapprehension on the part of Miller in 1866, which has been perpetuated since. The con- dition in which this species occurs during its first season has more than once led field-botanists to mistake it for ane “pani herbaceous Chrozophora tinctoria, or for the pros orm of that species, C. tinctoria, var. subplicata. The two tery are, however, readily distinguished when in ripe fruit because the capsules of C. tinctoria are red, those of C. spe fle: are blue; n in flower because the anthers of C. tinctoria are about 10 (varying from 9-11), whereas those of C. athens gifolia are is not ypc to maintain the variety proposed by megane in 1888 or the two varieties separated by the same author i 1899. Agerobens tase tinctoria, A. Juss. Tent. Gen. Eu ee 7, fig. 25/1. 11 (1824 Annua, he rbac ea; folia thomboi e0- vol tri- sgsiriovaty fibre lata vix longiora, margine repando-den- rubro-purpurea, Bistinets muricata. A widespread eae ar abet ct 5 ac of the Mediterranean basin, extending beyond Gibral thwards on the Atlantic seaboard to the Tagus a aed the ‘Boephovhs northward on the Euxine coast to the Danube and the Crimea. a. Zenuina, JJiill.-arg. in DC. Prodr. xv. 2, p. Erecta.—Heliotropium majus, Amat. Diosc. Enarrat. p. 43%, ; partim (1554); ibid. ed 2, p. 741 (1558). H. parvum, Clus. Hist. Nat. p. ult. cum icon. (1557); Lobel, Hist. SP | p. 133 Caahy- 7. aie Pinet , Hist. Pl. p. ages Ape (1561) ; acts ao p. 302 (1561); Camerar. Epit. Matt. p. L001 cum ‘icon. (15 86) ; id. Hort. Med. p. 73 (1588); a *Kratterb. p. 437a cum icon. (1611); Schmidt, Gesn. Op. Bot. t. 4, fig. 30 (1754). H. minus od et Tricoccon heron "Gesn. Hort. Germ. p. 261 — (1561) vulgare eine gallorum, etc., Pena et Lobel. Advers, p. 101 cum icon. _ H. minus tricoccum, Clus. Rar eee p- 395, cum icon. (i576): Bauh. Phytopin. Fi 487 (1596). Hy: ; A a pa nie Pp. Gi 1700); zdem, olio serrato — willoso, Tournef. Cor. p- (1703) ; a0 ree ye wigs _ 339, t. 17 (1712). Tournesol, oh, Magn, Nov. Char. Pl. p. 274 (1720); adabs. Fam. ii. p. 356 (1763). 104 Croton tinctorium, Linn. Sp. Pl. p. i o. (1753); Lamk Fl. Fr. ii. p. 198 (1778); id. Eneyc. Meth. . 212, var. B ety excl. (1786); All. Fl. Ped. ii. p. 47 ( (1787); a aay Hort. Re eg. Panorm, p- 406 (1789); Lamk Ill. t. 740, fig. 4 4 (1 790) ; Russell, Aleppo, p. 265, Sole ae Desf. Fl. Atl. ii. p- 354 (1798); Willd. ‘Sp: Phaave Ly! p: 8 (1805) : Geis. Crot. Monogr. p. 68 (1807); Ait. Hort. Kew = 2, v. p. 827 (1813); “agin oe oo - Bl. Gr. poh (1 Sl); ¢ Chaub. et Bory, Exped . Mor. oP ei pe: sa: Fi. (A Tournesolia os 188) nir. Hist. - Be 243 bra i); i . Med. p. ul re yst. Ve eg. iii. 850 (J 826) ; Spa "His. Vee; il 500 (1834) ; Reichb. Ic. Fl. Geet’ et Helv. t 4806 (1841) ; Mee. et Reut. Fl. Zante, pe 82 ea Not. Rep. Fl. Ligus. p. 366 (1844); Colm. Ca t, Pl, Catal. p ; ; 1. 2, pe , partim et quoad s es Taurid. eh: as FI. Da ea ee lll. p. 230 Seape Guch- Deli. (18 78 (1 92 (1904): id. Suppl. p. 9 (1908); Dur. et — Fl Libye. a 4 217 (19 my Pax 147. =P. 22, fig. 4 (1 912 ortucar. Estra madura: Setubal, Herb. Kunth! without precise locality, Welwitsch, 418! Alemtejo : Povoa. Welwitsch. oe Ses Silves, Welwitsch, 826! Serra do Arrabida, Daveau, 7! Serra do San Luiz; Vargem Valley, Daveau, 113! Spary. Andalusia: Xeres, Bourgeau, 443! Puerto Santa Maria t ! wan den Bosch! Cartama, Reverchon, 255! Granada, Rambur! Agua Blanca, Fimenes! base of Sierra » Bourgeau, 11821 1482! Willkomm, 372! Jaen, Blanco, 300! Mur urcia: He hae age iwee Valetivia: Benicarlo, Sennen, 1023! New Castile: M d, La Gasca! Reuter! Leon: Salamanca, Willkomm! Old Castile: Fontiveros, Willkomm. 105 Aragon: Saragossa, Ucheandia! Catalo Barcelona Butucriies! Compané! Llers a Hostalets, Sutin ‘ 16 8 | Balearic Islands: Majorca; Soller, Bianor, 1237! Minorca, Willkomm. ». Aude: Narbonne, Mertens! Pourret! Herault: Montpellier, Herb. Linn.! Herb. Paris! Gerard! Bentham | Viguier! Restinclitres, Bentham! near Laverune, Ramu! Val- argues, Ball! Gard: Manduel, Herb. rete Naudin! Nimes, Herb. Paris! Svezzol! Bouches du Rho near Marignac, Loret; Marseilles, Roux! Re iiaakaont” "i aaeeenid Aix, Grenier, Vaucluse: Avignon, Requien! Basses Alpes: Montfort, near Sisteron, Herb. Paris! Riez, Herb. Paris; St. “ ban, Site chon! Var: Le Laig, Huet 5 ‘Hanh in Herb. Schultes, 1139! Le Cannet, Hanry Herb. Billot, 3470 bis! Pierrefow, Chamberion in Herb. Billot, 3470! Toulon, Richard! Roquebrun, Bertrand! Hyéres, Auzandie; Grasse, erb. Paris! Frejus, Maire! Herb. Drake! Alpes Maritimes: Nice, de Notaris; near Antibes, Herb. Paris! Corsica : uglia, Salis! Dabeceau ! Tle Rousse, Bernard! between St. “Floren and Bastia, Soleirol, 3761! St. Florent, Mabille, 270 Iraty. Sardinia: Sestu, Maller! Mandas, Thomas! Cagliari, Barraud! Sicily: Palermo, Parlatore! Todaro, 432! Gussone! Ce Gnetiay ie Notaris. Roma Ro ome, Mond Pirotta! Campania: Naples, Regnier! dan d of aabienee, Bolle! Calabria : Pengtaho, Thomas! Gussone! Faventia, Magno- gati! Basilicata: without precise locality, Tenore; Pasquale. Apulia: near Manfredonia, Porta § Rigo! Abruzzo: base of Monte Marone, near Sulmona, Leresche! Pescara, Levier! Kuntze. Umbria: near Narni, Ball! The Marches: near Mace- rata and near Cape hae: Narducci! near Ascoli, Orsint. Emilia: near Rimini, Calde Avsrrra-Hunqary. Istria: near Trieste, Tommasini; Island of Lussin. Romer, 686! Dalmatia : Spalato, Pichler, 41! Petter, 123! Studnicka! Elfina, Bottert! Macarsca, Stossich! WLesina, Setter! Hohenacker, 281! without precise locality, Botticelli! Bosnia: Blagay, Knapp, 192! Monytenraro. Rijika, Bierbach! ALB Valmacu, i iter near Hagios Vasilios, Held- reich! Prevaes, Letourno Greece. Ionian Islands: Cephalonia; near Lari, Heldreich! nte, Margot; Ceri Speitz. Moraea: Sparta, Chaubard! Argos, Chaubard! Despreaus! Spruner! Napoli, Zuccarini, 442! Crete: Girapetro, era Cape — io Atropopoulo, 106 nides, 25! Heldreich, 442! 1189! near Kephissia, Heldreich! Locris; near Vitrinitea, Heldreich! Sporades; Perestri, Held- reich! Thessaly: Tyrnova, Sintenis. acedonia: near Lito- chori, Sintenis; Salonica, Nadji! Adamowicz! Turney. Thrace: Maku, Adamowicz! Constantinople, Noé, ai uLGARIA. Sadovo, Stribrny! without precise _ locality, Frivaldaty Russia. Crimea: Kapsichor, Callier, 197! Korbec, Leveillé Asta Minor. Bithynia: Broussa, Thirko; between Mardan and Broussa, J. S. Mill! Mysia: Island of Lesbos, Sibthorp; eet Calvert! oe Penk Se 146! without re locality Fleischer ! Bolin Cyprus. Nicosis an ad neighbourhood, Heldreich! between Cerignia and eps: Sintenis § Rigo, 624! near Omodos, Sin- tenis § Rigo, 627 Syria. Atsiibos Genkin, Peronin! Marasch, 1000 ft., Haussknecht! Aintab, 2000 ft., mire nay near Aleppo, Russell! Damascus: Haddat, gue? Chlora, Peyron! Aley, Peyron! Beirut, Peyron, 698! Pos Palestine: Galile ee; Shafa Hamr, Post, 217! Jerusalem, Roth! Meyers, 95 partly! Dead Sea, Meye rs § eh 395! Eeypr. Lower Eg Cairo and neighbourhood, Delile! Biconbbegs Raddi! Bove 374 partly! Samaritani! near Man- surah, Samaritani! Cyrenatca. Derna, Taubert, 629! Tunts. Zaghouan, Kralik! Djerba; El Kantara, Kralik! Doumet-Adanson § Bonnet! near Go Meee Letournouz ! Nabel, Gandoger, 18! Kerkenna Islands, Espina 5 eee 14. Algiers, Duriew! Berreau, Face Constantin, ris, 274! Bue wchinger ! Dunkerley! Sidi-ben-Omer, Romain! Philippi Choulette! Gambetta, Debeaur! Oran, Durando, Tlemen, Vignon! Biskra, Chevallier; 512! Kab: lie; Kar- at 2600 tt, Cn a Mount Magris, 3000 ft., Reverchon, 293! near Dellys, Salle, 142! Morocco. Tangier, Durand, 57! Salzmann! B. subplicata, Mill.-arg. in DC. Prodr. xy. 2, p. 749 (1866). Prostrata.—Boiss. Fl. Orient. iv. p. 1141 » partim et sp a Schweinfurth apud Kosseir lect. excl 79); Aschers. et Schwei Ul. Fl. Egypt. p. 138 1887) sa subplicata, Pax et K. Hoffm Pflanzenr. 7, vi. p. 24 partim et spec. apud Kosseir et in Afghania lect. excl, ( 1912) Eeyrr. Lower Egypt: ore ane neighbourhood, ie oat Bové, 374 ma inly! a aritan aq and Takrur, Schwein- furth A: ieee var. acai Schweinf. MSS.) ! typical variety, the ‘Tournesol,’ occurs in two so what “different conditions as regards shape of leaf. One of “ke two, which is the more usual, ‘corresponds with the plant from 107 plant trom Crete figured b Gesner, is me pies Soca in Greece, Crete and Syria. This latter form, i the leaves are lepidote petals render its separation comparatively easy. The prostrate variety from Egypt varies also in the shape and that would indicate this variety as being of the nature of a “te id. synonymy shows, writers from 1554 onwards have sxteatotied to identify the Tournesol with one or other of two plants described by Dioscorides and with at least one plant described by Pliny. Among modern eit Geiseler in 1807 accepted the identification of the species both with — Avorpémtov T) puxpov of the Greeks and with Heliotropion tricocco the Romans. The ini’ other authors to do this have been Pax and Hoffman, who in 1912 were prepared to accept the former but were doubtful about the latter determination: ‘ Diese Pflan they have remarked, ‘war schon den Arzten des hein: Altertums bekannt; sie is das 7Avotpdémuov puxpov des Dioscorides und " vielleicht das Heliotropon tricoccon des t heophrastus are few and scattered was only natural in remarks addressed to readers who must have been familiar with the plant intend Su s they are, however these characters are precise; none of them are incompatible with the identity of his kr athesahad with 7d ué Dioscorides After the simultaneous appearance in 1554 of the editions of Dioscorides by Mattioli tnd Castell-Branco the former published a trenchant review of some of his rival’s conclusions. It is tee that although they had come to different decisions as to the identity of *Avotpémtoy To wéya, no comment was made by Mattioli. There would have been ample justification for the criticism, had it been offered, that, as there is nothing xabatrep cxopriov ovpd about the inflorescence of the Tournesol, that plant could not possibly be #AvoTpémiov To uéya. Apparently the -_ nesol was a plant wn ‘ulate Mattioli was not acquainted, an this may account for his silence The identification by Clusius éf the Tournesol with HALOTPOTLOV 108 Td puxpdv has just as little to be said for it as that suggested by Amatus. ough the account which Dioscorides gave of pexpov is much shorter than that given of To peya, there is nothing in what Dioscorides has said which is inconsistent with fact that the account of Td puxpov is so brief, instead of justi- fine scholars, from the X VIith Century onwards; i in treating the nts as members of different natural families, suggests the li give oxopmiovpoy as a synonym of To pixpov as well as of Td péya. No doubt all four may be due to some error on the part o copyists; on the other hand, all four may equally well indicate that early students of Dioscorides were aware that’ 7d uLKpov, like to wéya, has an inflorescence ‘ curved like the tail of a scorpion.” That the latter is the more probable ec cuek ae is indicated by the fact that the great Vienna Codex supplies a portrait of Td jeKxpov which, although somewhat crude, un- doubtedly lepine & oe species we now know as Heliotropium supr- num, Linn, (Sp. Pl. p. 180). On ecological ero the suggestion of Clusius is ven less satiefactory “than that of Amatus , for the Tournesol at lea st does to Greek writers, it cannot have been the plant they knew as iNcrpimior TO [LLKpov. shee in 1830 (Comm. Diosc. 11. p. 642) accepted the hen aa ta tee, by Lobel in 1576 (Hist. Sting. p. 183) that the lesser Heliotrope “Of Dioscorides was the Tournesol, he was oe vertently led to believe that Gesner shared hie: pinoy What Gesner really did eo, was that the Tournesol might be the as) — a + ee Fe oO — oO 7D 3 fe | oO mee = =} + ky ° o ee, nae. o ha <-) QD 8 of ba} we Qu oO a “g f°) nm Sig ~ is] og 9 4 Qs eu i Q @O xt ed. by a as quite distinct and was identified by him with another plant. t will be noted that in the ‘Pflanzenreich ’ Pax and Hoffman, hat the Tournesol is also identical with the Heliotropion tri- coccon of a serious descriptive botanist in regardi ng any one of them as identical with either of the two eee a by Dioscorides. In both of the plants dealt with by P i 38 both it follows the sun during KH We are, however, con- 109 cerned here only with the one; which Pliny knew as Tricoccon, and although this plant also bore the name Seorpsucam, this was not because of its avOos émuxaurres Kabamep okopTiov ovpd but because ‘ semen ei est effigie scorpionis caudae.’ This fbb char- acter is in itself lee to render the identification ip nia by Gesner untenable. the applied botanist ua even stronger grounds for the See ‘f Gesner’s view that the Tournesol is the ‘ Heliotropium minus quod et Tricoccum sehen Toe hag not only on account of what Pliny has stated, but on account of what Pliny has left unsaid. The plant Pliny had i in mind was believed to possess the properties of a hi ifuge ; this alone should argh the least critical that his plant was not the Tournesol. On the other hand, the Tournesol is the source of a dye: had Pliny, who was nothing if o practical, been dealing with the Tournesol, it ii nceivable to the economic botanist that this outstanding character should i passed over 8. Chrozophora hierosolymitana, Spr. Syst. Le: ill. 0 1826). Annua, herbacea, erecta; folia ovata sak su Pohotaboiiink: quam lata vix longiora, margine distincte dentata vel subintegra, utrinque dense vel laxius velutina ; antherae 4-5, raro 6-8; capsula matura rubro- -purpured, distincte muricata. i Crbtn Babtoeiem: Russell, Aleppo, ii. p. 265, partim tieek non Linn. Croton pli- catum, ‘Sieb. Avis. Herb. Palaest. Pp: 7 (1821); non Vahl. Croton oblongifolium, Sieb. ex Spr. l.c. (1826); non Del. Croton ne ve Wall. Cat. 7716 G (1880). Chrozophora Sieberi, Presl, Bemerk. p. 109 (1844). ico aeel re Ledeb. Fl. ie : . So Bot. xix. p- 186 ass); B son f. FL. Brit. Tnd. vy p. 408 quoad spec. — lepidotis tantum (1887); Watt. Dict. Econ. i od. Ind. 21 (1889) ; a Bulletin, 1889, p. 279 et 1896, 233 ; Tigiky, eee Hort. Tift. iv. v. [FL oe - mg sera § Radde. Mus. Caucas. p. 154 (1901) ; ates et K. . in Engl. Pflanze . 147. vi. p. 22, partim et quoad s ra r otiaslbers 1270, Sintenis 1462, Becker, Schlangintweit 2624, homso (1912); non A. Juss. Chrozophora tinctoriae Adr. de Juss. ? affinis, Bidigs Rel. spas P- 35 tna et ee Acad. Pétersb. Vins, ps eee 854). lia, Baill. Etud. gén. quoad ole. Scind. 1B 4 n a. Juss. C Vivaesaiio subplicata, Pax ok Hoffm ( HN "24 4 partim net quoad sp. afghan. [Griffith] tantum (1913). his species is characteristic of and is narod spread through- out North-west India, the North-west Himalaya, Baluchistan, Afghanistan, Persia, Turkestan, Georgia, Mextued Asia ates 110 Mesopotamia, and Syria, with an isolated outlying centre in Western Arabia and another in Egypt. Chrozophora hierosoly- mse obliqua of which it has the anthers and the et esd : ie when C. shinny te has the pannose pubescence o sec the arger, more muricate ca sules. In the Pawjab this plant is sometimes collected and used as fuel sSeomnat rt). Inpii. Upper Gangetic Plain: Muttra; ees Hard- wicke in Herb Wallich, 7716 G! Panjab: Sirhind; Ambala, Edgeworth, 127! Firozpur, Thomson, 1473! Bari Doab; Am- ritsar, T. Anderson, nea Lahore, Thomson, 1473 bis! Branilis, 2040! Rechna Doab; without precise locality, Thomson! Jech Doab; Shahpur, Thal Deaeee 900 ft., Ram Baksh in Herb. Drummond, 6347! Sind-Sagar Doab; Salt Range, Aitchison! Ahmadabad, 900 ft., Naran Das in Herb Drummond, 6344 ! Rawal Pindi, Aitchison, 531/1092! Schlagintweit, 10945! Camp- bellpur, Stewart, 18! North-west Frontier Province: Waziristan ; Tonk, Duthie, 7205! Gomal Pass, Gage! North-west Himalaya: Kashmir; without precise locality, Thomson! Gilgit; Kagushi, 6500 ft., meee 368! Chitral; Warai, 4500 ft., Ontacrs in Herb Duthie, "17534 Batvcwistan. Quetta: Shela, 4800 ft., Lace, 4089! Teree, near Mastung, Stocks, 1079! Fort Sandeman, ages 19007 Mekran: on the south coast at Kapotcham, Pierc ArGuanisTan. Kandahar: near Kandahar, Griffith, 681 (K.D. 4792)! Helmund River, Aztchison, 70! 732! Kabul: be- tween Thal and Kurram, Aitehison; 425! Herat: Hari-rud Val- ley, Aitchison! _ 'TURKES TAN. Samarkand: between Bokhara and Samarkand, Lehmann; near Samarkand, Fedtschenko! Khorasan: near Meshed, Bunge! Luristan: without precise Tocatity, Rivka, — 5297 ! Arabistan : Mohammerah, ! Luristan: Chrysan Valley ange Trak-Ajemi Ispahan, Awoher, 2008 ! Teshant “Glov St. John! Teheran, Casson! Persian Kur distan: near Avro cree Haussknecht! Azer- bijan: near Khoi, Szovitz, 450! Caucasta. Daghestan: ee d, Becker, 72! 73! without pene locality, ‘Steven! Georgia: Tiflis, Radde, 253! Schu- n! Metajowski! Mugah, Radde, 328! Helenendorf, Hohe- I Sl Asia Min Pontus: Amasia, el ata 1270 (issued as C. tinctoria) 1770 (issued as C. verbascifolia)! iil Mesororamia, Turkish Kurdistan: Mardin, Stapf, 1462! Upper ef ety Montbret! Mossul: near Mossul, Kotschy, 441 partly! Bauere! bane che Bagdad: near Bagdad, Schlafli, 80! Kapp, PAE, 2007 Syria. Aleppo: © bd 3 a sm Qu ® © pat fen =] 05 4 = i= t+ Bb © re) 3 ca) | B ao mn = both hard- and soft-shelled nuts in the same bunches. It was further noted that the bunches of nuts yielded by this variety were much smaller than those of any other forms. The following particulars are given of the yield in bunches of fruit from the palms planted in 1912 :— Abetumtum :—9 plants, 11 bunches, average weight 7 lbs. epa :— 17 plants, 19 bunches, average weight 5 Ibs. Abobobe:— 19 plants, 19 bunches, average weight 3% lbs. All the bunches were very small in size. : Before any answer as to the true breeding of the different varieties can be given, it is clear that more careful experiments must be devised. é In the first place hand-pollination of a variety with pollen from a plant of the same variety or from the male inflorescence of the same tree must be undertaken. No reliance can be placed on seed promiscously collected from any particular tree since cross-fertilization of trees of different type is constantly going on and the resultant offspring from such crossed trees will exhibit i is a in the Report of the ‘‘ Committee on Edible and Oil-Producing Nuts and Seeds,’’ 1916. Cd. 82-47. CULTIVATION. ; The oil palm in West Africa is, as a rule, subjected to no special cultivation and it is therefore all the more interesting to learn from the Gold Coast Report that palms subjected to proper cultural conditions have responded by a marked increase in the P : : has been cleaned, the palms being thinned to a reasonable dis- tance apart. In aidition , there are odd palm trees scattered about 122 the station growing in land which is under definite cultivation for other crops and the marked effect of good treatment on these trees is shown in the following table. No. of bear- No. of Weight of | Weight of ing Palms. | bunches. bunches. | clean fruit. | Tbs. Ibs. b-acre plot... sez 357 223 4,355 2,737 Odd Palms... vas 224 805 | 13,071 8,217 In both cases the yield is considerably in excess of the previous year, the return from the scattered palms being quite remark- able. It should be noted that the 5-acre plot has not yet been under soil cultivation. il palm cultivation is being taken up outside West Africa and promising results are indicated in the Agricultura] Bulletin, Federated Malay States, vol. v., Nos. 11 and 12, 1917, 439. Several acres of Elaeis guineensis were planted at the Kuala —— experiment station in December, 1912, and a number of the ditions are suitable. n this connection the results obtained in the Seychelles with the cultivation of the African oil palm are 1 as is shown by the following account taken from the Annual ee — Agriculture and Crown Lands, Seychelles, for the year “The results hitherto obtained with another introduced palm v be large extent, as already recommended in 1912. This palm, which produces several i and palm kernels, has foun unch was as much as 2392 and each fruit was fully developed and perfectly ripe; a bunch of this size yields as much oil as fruits are so good to eat that all the emp oyees and labourers of the Botanic Station are very fond of them, besides the Ashanti political prisoners who are regular visitors of the garden in search 123 will set out a large plantation as soon as possible. Numerous small plots have been planted out on the Crown Lands. In West would, besides, produce a fruit of no small dietetic value for the poorer inhabitants.” 124. lishment there of plantations of = palm similar to those which are understood to have already ~been established in various Malayan colon both English at Dute yale n vee e of other aes derived, as ce dae of the oil palm n Africa are, from the saree ‘the a lishment of plan- tations, with their more satisfact economic con ditions, has 1 in- eae conditions, this important tree may cease to be a means of livelihood for their subjects and a source of revenue for themselves. IX.—MISCELLANEOUS NOTES. Mr. Donatp Ha.xerstoy, a temporary Sub-foreman in the Royal Botanic Gardens, Kew, has been appointed by the Secre- tary of State for the Colonies, on the recommendation of Kew, an Assistant eens gla al Officer in the Department of Agri- culture, Ugand Gaston ALLARD, OF ANGERS. —Harly i in January the famous Arboriculturist of ‘Angers—M. Gaston Allard—died at his home, La Maulévrie, aged nearly 80 Mesias The writer is almost cer- ear to France and bs er in suet. 1914 but Tie Salbreak the put an end to the p projec n the ea rly part o Oetober whils visiting er Allard’s Reigkbott the Marquis de Charnacé, I was enabled to carry out this pla I found M. Allard sitting with a friend on a seat in the garden on me sunny side of his pleasant country house, and after tellin im I had come from Paris to see his famous trees he insiste to appreciate; in the ca case of the latter - — afted the male and s he was especially 1 125 He travelled eananey for three years in Algeria, Tunis and Morocco, when he made the flora of those countries his especial study, at a time when travel there was a matter of much diffi- cog a some seat n to prepare ‘the ground for the Arboretum io Angers in 1858, and the fret plantings were made in 1863. To-day the Arboretum contains a full collection of eed a Nach ae China, Japan and the Mediterranean litto He has bequeathed this collection of half a century to the Pasteur Institute, and it is expected that his house will be used by the Institute as a laboratory of vegetable biology. Doubtless he felt that ihis was a more certain way to ——— it to posterity would wish for it R. S. BALFOUR. In this country we ete probably only one person, the Earl of Ducie, who, in the success es extent of his plantings and the lengt of time over ‘hte y have extended, has a record Sa that of M. Allard ard never advertised his successes dee ¥ is perhaps due to Mis modesty that relations between aS d Kew were not established until late in “his life. In recent years his collection has been visited by officers of this mticleeent and they were: always welcomed with charming courtesy. Botany y. She ec vied on “8 first ie researches in the 126 Miss Sargant was an exceptionally skilled observer, and her work was of the highest degree of accuracy. Her style in writing her papers was as clear and vigorous as her observation was exact. ' ‘cogD) s « Pe Boissier Herbarium.—A letter received at Kew from Professor Chodat contains the important announcement that the famous herbarium accumulated by Mr. E.. Boissier, after his death the property of his son-in-law the late Mr. W. Barbey, along wi : : : e ary last, to the Botanical Institute of the University of Geneva. The collections and the library are to be transported to the Uni- versity. Mr. Beauverd, the present curator of the herbarium will continue in office after the transfer. Himalayan Exploration.—A valuable contribution to our knowledge of Himalayan geography, and at the same time a graceful appreciation of Sir Joseph Hooker’s services in the same field, has been published by Lieut.-Col. W. J. Buchanan, I.M.S., C.I.E., in a paper* which appears in volume xiv. of Bengal Past and Present, and of which the author has sent a reprint to Kew. m . ope nag preliminary to a brief biographical sketch of Sir f S ce J accurate as a guide-book that it has been my custom for some = 3 ae the few mistakes which he made, as discovered y othe discussio n. includes the world’s loftiest mountain. *“Tn the Footsteps of Hooker through Sikki 2 A vinien Ain. [Onlouin IA 127 scribes more particularly what is believed to be Makula which, from his point of view, was the more conspicuous. Lieut.-Col. Buchanan’s paper is provided with two photographs, one of the Everest-Makula peaks and the other of Kinchinjanga, Kubra and e northwards to just beyond Mombasa. It is not, h confined to the actual coast zone. One locality, recorded by Dr. characteristic species is i nu indicate. In his “Travels in the Coastlands of British East 128 Africa and the islands of Zanzibar and Pemba,’’ published in 1898, Mr. W. W. A. Fitzgerald refers on more than one occasion to the occurrence of a Cycad. Although Fitzgerald was aware of possibility, however, that ‘Mikindu ’ may prove to be EL. Hilde- brandtii and it is hoped that this note may induce visitors to the island to transmit to Kew the material required to settle the doubt. With the references to mainland localities in Fitzgerald’s work we appear to be on surer ground. In his summary of the notable vegetable productions seen Fitzgerald gives (p. 617) “ ‘ Kitsapu, a Cycad. The heart of this palm-like tree is cut into dice and subjected to soaking and drying for three days, after which it sage contributed by Taylor to Fitzgerald’s work there is a refer- ence to this Cycad which indicates that only one species is to be met with in British East Africa and that species, as the various specimens collected by Taylor show, is E. Hildebrandtii. f Fitzgerald’s own references to this species do not contradict this conclusion. In a locality 14 miles north of Rabai, whence between Mombasa and the mouth of the Voi River, Fitzgerald p- 288). Again, somewhat further north, between M’Tondua and Konjora, in 3° 28’ S. Island, in 1° 55/ S., 41° 1’ E., his attention was c o a large wild Cycad with an enormous pine-apple shaped fruit which the i .p. 425 Kew Bulletin, 1918.] Mal by& Sons, Photo-Litho- y . it Microcoridia of Botrytis cinerea. ~ age 123) Te face p /} [Crown Copyright Reserved. ROYAL BOTANIC GARDENS, KEW. BULLETIN MISCELLANEOUS INFORMATION. No. 4] [1918 X.—THE MICROCONIDIA OF BOTRYTIS CINEREA. StTupDIeS FROM THE PatHotocicat Lasoratory VII. Witiim B. Brrerwey. Introduction.—In 1887 Iindner* observed that the cell con- tents in old hyphae of Botrytis cinerea occasionally contracted earing a chain of minute conidia at their tips. An excellent delineation of this microsporogeny 1s k. which are a large num is equally abundant. layer; the nuclei are : pheral cytoplasm and appearing as sma The portion of the filament - tains an accumulation of metachro enter the sporidia through their mother ce s e nuclei — difficult to distinguish during the formation of the sporidia, but * Li : : . 1]. ¥, 1887. BE ne Et tenes “4 Onc Rend. du Lab. de Carlsberg. v- + Lorrain Smith, A.: Journ. Bot. xli. 1903. J uncou.t eA pon cei A.: Centr. f. Bakt. ii. x. 1903. (5353.) Wet. 196—794. 1,125. 5/18, J.T. &S., Ltd. G.14. Sch. 12. 130 it would appear that only one aoe in each sterigma; each of the sporidia also contains a single nucleus, this resulting from the division of the nucleus in the eiatiraia the formation of the w settee a appears, therefore, to be identical with that of the nidia of Penicillium glaucum described by Dangeard and Gué en. The mature sporidia each contain a single nucleus, situated at the periphery, and a vacuole occupying the greater part of their volume; one or two metachromatic granules are also to be observed in the ‘peripheral layer of cytoplasm. ““In this medium (distilled water) the osmotic pressure of which is very feeble, many curious morphological phenomena are Istvanffi*. Microsporogeny only occurred when the fungus was grown in glycerine, and the foll oan characteristic forms were xhibi poston resemblance to a promycelium (type Sclevotin ta); (b) 8 Patem on the sterigmata in the shape of a sickle (faucille) ; “cel arge masses of the type gamo- meterephal-eeneua Ns (d) little chains at the tips of fila- IL the piled be authors regarded this type of spore aad as a very rare and abnormal or bizarre occurrence, and in no case was the deve oye of in microconidia obse In 1908 Brookst published his ‘‘ Observations on the Biology of Botrytis aad aay and found that microconidial formation oceurred when the fun ngus was cultivated upon bouillon with 10 per cent. gela i He states that in this medium ‘‘the growth was comparatively feeble, and during the first puree generations red conidiophores bearing microconidia were bor In the urth generation of the fungus upon the b bousllon medium the Browt was more fearant and the normal conidiophores were pats uced. It is not known whether this senge 1 in the mode of reproduction was dependent upon some alteration in the condi- tions of experiment, such as accounts for the different modes of reproduction in some algae, as Klebs has shown. If this were the ease the alteration must have been a slight one, for the cultures were kept in a room where the obvious physical ga controlling the growth were the same. It may be that the fungus, having become scene’ during three previous generati to the bouillon medium, was sufficiently invigorated to produce again the ietoia ao of conidiophore.’ * Istvanffi, Gy. de,: Anns. de !’Inst. Centr. Ampel. Roy. Hongr. iii. 1905. tt should be foe that raechictg the whole of his work ead ee ‘Botrytis cinerea as identical with Sclerotinia Fuckelian. In view of the very ba discrete eae without genetic relationship, and belonging 7 widely se separated grou See Li nd, J.—Danish Fungi in the Herbarium s nu p, Copen! agen: 1918 Pethy brides: G. H—Journ Dept. Treleed’ xi. No. 3, 1911; xvi. No. ‘4, 1916; Smith, R. E.—Bot. Cae. Xxix. 1900. + Brooks, F. T.: Ann. Bot. xxii. 1908. 131 In 1902 or mca Ce capo a remarkable account of the on morphism of a species of Botrytis attacking Salvia Hor mmum, L. His cualales eke be summarised as follows: —The senate: which the author designates Botrytis hormini, gives rise (i.) a sterile mycelium maieplyeg by fragmentation and producing the forms vi., and vil. (ii.) the type Polyactis na reproduces by conidia (ui.) the type Cristularia with microconidia, which never reproduce the Cristularia form, but the type ii. (Poly- re and the type vii. (Gamocladocephalo-merizo- a ) ak e sclerotia, (Eyre sonidisl ‘form of the type Alternaria which gives rise to simi ores. _(vii.) an andi conidial ape Gamocladocephalo-merizo- sporica having two kinds of microconidia. The first correspond to those of the form ii. (Polyactis) ibe give rise either to a similar type (Colyact) or to the form vii. (abnormal). The second of the type iii. (Crate. laria) ee ped a io Rie type ili. (Cristularia) or the type ii. (Polyactis). It is very unfortunate — peace” all experimental details are omitted from this a his polymorphism is so Peatnané and of so unusual an order that until some degree of confirmation is forthcoming the author’s interpretation of his results must ne accepted with much reserve. n examination of the many fi ee ites ying the memoire makes it appear very probable that ch es i. (fragmenting sterile mycelium i. (Polyactis), lii. (Cristularia), iv. “acne vii. mer rious phases of the common fungus Botrytis cinerea: whilst v. aha vi., the Macros- portum-Alternaria types, were merely different growth phases of the common contaminating fungus Macrosporiwm sp. The latter supposition receives strong confirmation from the fact that on ne rm represented in plate xx. of his work. erhaps no single fungus has baat so thoroughly aig by so so Many competent observers as Botrytis ct a, OF is centre of such an extensive literature; and yet only in the six or = Parneti, R.: Atti, del. R. Ist. Bot. d’Univ. di Pavia, vii. 1902. a2 ~ 132 contaminating fungus Penicillium sp. In consequence, there is among myco very general scepticism as to the reality of any genetic relationship between e crosporogenating mi y and Botrytis cinerea. Even if the evidence of such relationship be admitted, the microconidia are merely regarded as @ bizarre form without morphological status, and are un- a Wwitne, in systematic treatises on the fungi. hilst carrying out an extensive series of pure culture experi- ments with this fungus, the formation of microconidia was noted and the opportunity was taken to attempt a decisive settlement of the question of ihe true status of these spores. The process of microsporogeny was first observed in a culture- six weeks old of a strain of the fungus derived from onion leaves Bb 9 B B o a ss °o B 4 a Es > ® S i=) S$ ©. Ry & cillioid gro c mycelium. It was at first suspected that the former was develop- ing parasitically upon the latter, as in the case of Cicinnobolus upon the mycelium of Hrysiphe,* or certain Penicillia upon the that the formation of minute conidia is a true developmental Formation or Mrcrocontpra. The microconidia may originate either from the vegetative mycelium, from the cells of the conidiophores, or directly b the germination of the conidia. The latter as the simplest case will be described first. . * See De Bary in De Bary, A. and Wordnin, M.: Beitrige zur Morphologie- und Physiologie der Pilze, 1870. + Brefeld, O. : Untersuchungen ii. 1874, 133 ~vacuolate. When a length of 10-15 » has b i i - tt growth usually ceases, a septum ining oF sia: ant ber a (loc. cit.) was of this nature, for true budding in the absence of a definite sterigmata has neither been noted by other observers nor found during my own work. The origin of Microconidia from the conidiophores.— The for- microconidia from the conidiophores presents man differently from the upper region, and it will conduce to clear- néss if these are treated separately. Upper region.—The cells of the upper region are thin walled cand densely filled with usually homogeneous or finely granular = TBS Ge PO eo * Brefeld, 0.: Untersuchungen ix. 1591; xy, 1y12. 134 protoplasm which by virtue of its sporogenating function is in a vigorous and plastic condition. : en microconidial formation occurs the spore-bearing branch- e ise to lon slender hyphae which either terminate in slightly swollen cells. from which arise the sterigmata (Pl. v. fig. 8), or more rarely pass directly into a single sterigma. Not infrequently these hyphae are themselves branched, so that a freely ramifying system is formed, producing an immense number of spores. ee or branched sterigmata often arise laterally on these requently the sporogenous branches of the conidiophore do: not proliferate, but slender hyphae develop from immediately below them, and these originate the microconidia as already described. Microconidial formation may occur in one of two ways. figs. 9 an ). Not infrequently several old cells are traversed by the sporogenating hypha before the principal formation of urs, and often the containing cells are ultimately so ‘ | E . j h on germ tion give rise to microconidia. Often this contraction does not occur, but in all cases there is an apparent renewal of vigour and a marked increase in the quantity of cell content. et pnaearising from of the vegetative mycelii 3 either di . ycehum may give rise to sterigmata borne e mycelium of various species 135 Geminella* aud Sclerotiniat and are known in certain other fungi. When large clusters of sterigmata occur in close proximity they produce the G ladocephalo-merizosporica and Cristularia forms of Farneti. Not infrequently the rejuvenescent cells germinate into adjacent cells and give rise arge numbers : sai : fi a single chain along the lumen, simulating the appearance of endoconidia (PI. v. fig. 8). In all cases, however, microsporogeny takes place by a process of acropetal segmentation from a spe- cialised sterigma and never by fragmentation} or budding. or by following the actual development in any single case. septum is formed, separating the spore from the sterigma (PI. v. fig. 11 g). ‘This septum splits along its middle lamella liberating — neck (Pl. v. fig. 11). The swelling increases, is separated as be- fore by a septum which splits in the plane of the middle lamella freeing the second snore (Pl. v. fig. h, a, b, c, d, e). is process is continued so that ultimately a long chain or cluster of spores is formed on each sterigma, the new spores pushing ee ¥ ceedingly thin pellicle, and contain than one, central and highly refractive granule of an amorphous he recent figures of Gilbert, A. H., and Bennett, C. W.: Phytopathology vii. 1917. i i ial mentation may occur, but the tterly different from the microconidia. stat direct response to a Ss sion : oidia eck, ea al Be germinate under favourable conditions giving rise to the normal vegetative mycelium, and their pr Hebei method of tiding the fungus over a suddenly arising ad E mentation hae been recorded by Berlese (Malphighia 1889), Farneti and Istvanfii. . 136 spore contains abundant glycogen and a single nucleus which is not easy to distinguish. The sterigmata are also uninucleate and contain glycogen. GERMINATION OF THE Microconipta. On germination which occurs in from twenty-four to forty- eight hours after immersion in water or a nutrient medium in his cultures on bouillon that the first three generations were with distilled water, and th i > at 22° C-23° ¢ of the laboratory at a temperature of 15° C.—17° (. _ One of the * Meyer, A.: Die Zelle der Bakterien, Jena, 1912. : + Wager, H. H., and Peniston, A.: Ann, Bot. xxiv, 1910. 187 cations were formed in from ten to fifteen days. ‘exception the control plants remained perfectly healthy. Inocu- lations of unwounded surfaces were a total failure. Nothing resembling the polymorphism described ‘by Farneti nor the 5 imorphism obtained by Brooks was noted at this, or any other stage, of the investigation. Th croconidia appea be ely a normal developmental stage in the life cycle of Conpitions or Microconrprat Formation. In 1896, Klebs* demonstrated that the developmental stages in the life-history of the algae and fungi are determined by specific quantitative changes in the conditions of life of the organisms. seemed desirable therefore in the present case to ascertain if not the temperature, was the primary factor involved. These authors found the sterility to become permanent, but this has not - * Klebs, G.: Die Bedingung der Fortpflanzung bei Algen und. Pilzen. Jena, 1896 ; also Jahr. f. wiss. Bot. 1898 and 1900; Probleme der Entwicke- lung iii. Biol. Centralb. 1904. t 138. conditions xposure to very intense li¢ht is inimical to myce- lial development, and the fungus grows feebly downward into the substratum. , however, the intensity of light be decreased 2: r which microconidial formation has occurred abundantly have been maintained either in the diffuse light of the laboratory, or in the darkness of an incubator. Varying intensity of light there- fore does not affect the size of the conidia nor the type of sporo- geny, but only the amount and rapidity of conidial formation. The same holds true of light of different wave lengths. Exposure to the less refrangible rays of the spectrum inhibits the formation of conidia, whilst these are produced in abundance in blue or violet light. The quality of sporogeny, however, is not affected, but only the quantity; either Polyactis conidia are pro- duced or there is sterility*. Light cannot therefore be the de- termining factor in the production of microconidia. sterile form with proliferating conidiophores “forme inter- ty of vegetable media the fungus produces spores very abundantly between the temperatures of 16° C. and 25° C. With raised or lowered temperature mycelial growth and spore A gor and the former possessing a slightly greater range. With vary- * See Costantin, J. : Bull. Soc. bot. de Fr. 1889; Klein, L.: Bot. Zeit. 1885; Reidemeister, W.: Ann. Mycol. vii. 1909; Moreau, M. and F.: Bull. Soc. bot. de Fr. 1913. 7 + Beauverie, J.: Na hr Rend. 128, 1899, and 133, 1901, also Anns. Univ. ty dona Nouv. sér. iii. uverie, J., and Guilliermond; Centr. f. Bakt. ii. ‘ 139 ing temperature there is no 0 change in the size or ~~ of conidia. Cultures which roduced microconidia in pro- The factor cf temperature, therefore, like those of light a nd humidity is not the determining condition in the fortixtion- of microconidia Nuatrition.—The factor oa ached is much more complex than the three factors on have been briefly commented upon. Many distinct influences are com enc such for example as the physiological avedlability of the nutritive substance and its accessibility to the fungus, the dilution or concentration i Hy © B Ru 5 6 o nD mn 3 5 - th aS mn . Failure to resets seh beteen thes component factors renders invalid the results obtai in perhaps the majority of investigations on the nutrition of fun The substrata on which microsporogeny has bedtierél in the present work saietw the following :—Potato agar, potato gela- tine, parsnip agar, parsnip gelatine, Quaker oat agar, Quaker oat gelatine, steamed slants of tuber of potato and artichoke; roots of red beet, sugar beet, carrot, turnip, and swede ; fruits of lations into the following living hosts have given microco- f daffodil, and onion; fruits of apple, pear, quince, and tomato; roots of beet, sugar beet, carrot, sw and turnip; and tubers of potato and artichoke. penta! Cale op has been observed on the following naturally infected hos Bulbs of onion and hyacinth, a s of apple, tomato, and wera leaves of cabbage, densutte cotyledon, clematis, geranium, and other herbaceous plants, and ne of carrot. It may also conve- niently be noted here that strains of Botrytis derieea from the following original hosts have produced microconidia in culture 0, r primula, shoots and ‘“‘fruits” of fig trees, ada “the st ee of 3 an Aesculus rae which had bas killed by the fun authors have oe bel cay upon peptone, a tlled ‘water, but in the present coal this hypothesis is very improbable, for, as noted above, microconidia occur regularly on all kinds media and natural hosts under perfectly normal conditions, oak possess a well-defined bi bea in the life- -cycle of the fungus. © F rogeny is a ap pein or be ound * Since sg hae was written microconidia have been found in strains of the fungus derived from thirty additional hosts, and have incurred im are on all I the media and natural substrata + See Worsdell, W. C.: Principles of Plant Teratology, London, 1915. 140 a gradual anatomy between the usual pean form upon a epic a edium, the microconidial form upon an unsuit- edium; and if iad be possible to ohiess every stage in this transition by growing the fungus upon a carefully graduated series of media. Experiments have shown that this is not the needs. Moreover, the formation of microconidia of constant size by pach jinetie of very specialised type upon a penicillioid form of conidiophore is a development perfectly distinct and abruptly aeperatad ‘hon the Polyactis fructification: there are no transi- tional forms of either spore or conidiophore. Numerous studies have been made* on the effect of various food conditions-upon spore formation i ytis cinere production, or accelerate or retard the formation of spores ne do inca hae pre like physiological repletion controls the quantity an ¢ the quality of conidial formation,t and cannot, therefore, be ne eae factor in microsporogeny name ils others produce these minute spores elium principally from the old myce 142 quickly reduced to a semi-fluid or pulpy condition. This rapid disintegration is accelerated by innumerable saprophytic fungi operation the host plant has either completely disappeared, or exists in such a condition that few mycologists care to retain it j i n th fructifications. Again in cultures of the fungus on artificial media micro- sporogeny does not often occur within a period of approximately one month. i ut t r from these the microconidia germinate and give rise to a normal mycelium producing the usual conidiophores, so that the microco- nidial origin of the new culture remains unsuspected. If b chance ion i e wths. It would appear, therefore, that in these cultures Rear formation pies not 143 related to physiological starvation or any direct nutritional condition; but to some accessory factor. Istva nf obtained microconidia by germinating conidia in gage orhedre of = chieorse (1 per cent., 5 per cent., 10 per nt.). The mycelial development was sparse, but microconidia attest preleacte; the sterigmata being bor ne often on highly sheer hyphae. The author remarks: ‘‘ In these old cultures or three weeks) we have found the best sterigmatous va sedation! ” Although Istvan Spree out a very great wiih germination experiments with the spores of this fungus using a large number of different media, many of an obviously physio logically unsuitable nature, only in this particular case microsporogeny noted, and it would appear improbable thersfote _e i abnormal sporogenous eesaaied was related to the uy that in this pabiccaiban 6h case Istvan may have chanced to use spores from an old culture when, as has been shown, they would normally pcan in the manner > denseted by = ee microconidia were borne. In the aette generation of the fungus upon bs e bouillon medium the growth ¥ more luxuriant and Penaration nde sped it “fol ow the co found by the French authors and form microconidia at first and later in the same medium gradually change the quality of its spore fo Furthermore it has been ‘pointed out that probably c. cal, difference in the media which in the first three pais ne with the normal controlling factors. Among my own cultures premature microsporogeny has oc- curred in the duitowing cases, always, however, pat preceded by the normal production of Polyactis conidia. strain from onion bulb growing on steamed carrot formed sihaebeoaiaich after fourteen days; a strain from ntirrhinum on potato ager after eleven days, and a similar culture after fourtee strain from lettuce ihr on su t root pat microsporogeny after thirt n days. In each of these cases the culture was vigorous and coacpandal production at first very 144 scanty, In exactly similar ness of the same date growing Maga identical conditions microsporogeny only occurred after ur weeks’ time, peck: the peemeeire sporogeny is most stages reversed or 0 This résumé of Ges: md cases of premature microsporogeny in the formation of microconidia. in it may ointed oat that the results oiicined eae pee saihend ne not sorttuda with each other or with those either of myself or of se. investigators* and it would appear that in these abnormal cases the usual life stage—(with a sclerotial stage according to nutritional con- ditions) has been issn & interfered with and the spore stages reversed, or even elimin This epeceice eat ot Tite ‘eyele i is by no means a rare occur- rence in the vegetable kingdom. Perhaps one of the most strik- ing cases is that ‘deter ibed by Mébiust of the oak in which almost certain developmental stages in many elk s ants orna- mental s ry subjecting them to specific quantitative changes of environment, such as temperature and humidity. § he case technique. Thus Klebs has shown that ‘‘ by modifying the external conditions it is possible to induce fungi to grow con- tinuously for several years or, in the course of a few days, to die after an enormous production of sexual or asexual cells. In some instances even an almost complete stoppage of growth ma be caused, reproductive cells being scarcely formed before the organism is again compelled to resort to'reproduction. Thus the sequence of the different stages in development can be modified as we ma The reproductive stages in the life cycle of Botrytis synchronise with crests in the rhythmic working of the metabolic changes in the hietteee When these changes proceed eager ted Ae the e for ee Behrens ; Smith, R. E.; Istvanffi; Reidemeister + rete (loc. MAE There vs no io evidence for the connection of Botrytis cinerea with t ewer ? Beitrige zur Lehre von der Fortpflanzung, Jena, 1897. § See Molisch, H.: Das Warmbad, Jena, 1909. 145 which is is aha in the formation of Pe conidia. This g occurs is represent time hie days. Put briefly, the fungus has, in a few aie 8, P dined to a physiological condition which is usually only récit after a period of about four week at which pe meet occurs is re sated in time fet of several —o t by specifically ee vo ga con- ditions and here the Sakedtering factor is of a cal nature* he life proved’ are so accelerate that ee same Soh yecdagegs age — e reached in the course o I opimaixily controlling microsporogeny is apparently one of age, and it would appear highly probable that in a a ent u 0 x interfering ae saages in the medium. Both meat bouillon and the earlier peptone media are of very epi Spa and little known. chemical deinpentr, ri these ab- atop Be ete einere Sractihogiion and other con occur until the fungus is about one mo The importance of this process in the life history of the fungus is that it provides a means whereby the eae when under ‘conditions unfavourable for dissemination and gro may pro- ce immense numbers of a type of spore so Foteate and light * See Ward, H. M.: Phil. Trans. Roy. Soc. Lond. B. 191, 1899, and Brierley, W. B.: Ann. Bot. xxxi. 1917. 146 that its rapid spread is certain, and in consequence the continuity of existence of the fungus is ensured. I have received much assistance in the preparation of cultures and slides from Miss M. N. Owen, Temporary Technical Assistant in the laboratory, and I am glad of this opportunity to record my indebtedness to her. EXPLANATION OF PLATE vy. The figures were drawn with the aid of a Zeiss camera lucida. A Swift } objective N.A. 0-62 and a III eyepiece were used for gur j Figs. 1—6.—Normal conidia of a strain of Botrytis from Primula leaves germinating directly to form microconidia. type noted in 6 Fig. 8.—The extreme tip of one of the filaments arising conidiophore. The terminal cell is swollen and gives rise to sterigmata. Within the filament is a sterigma which is produc- ing microconidia forming a single chain along the lumen. Below these are seen two cells with irregularly thickened walls, which function as chlamydospores. . 1g. 9.—A rejuvenated cell which has protruded through the pore in the transverse wall and formed a single sterigma produc- ing spores, within the filament. a and b show the development in a of eleven hours. Fig. 10.—A more complicated and usual case of the formation seen in 9 Seacrest NT 147 XI.—TAXOTROPHIS AND BALANOSTREBLUS. J. Hutcurson. : » genus Taxotrophis (Urticaceae) was established by Blume in 1852, and included two species, a Javan plant, 7’. javanica, Bureau* based a new genus, Phyllochlamys, on 7’, Rowburghi. dn He Thwaites added T. zeylanica, founded on Epicarpurus sca ca, Thw., and in 1886 and 1913 Vidal and Elmer Bi ge on Streblus ‘macrophyllus, ume, from the Celebes. sg Gamble described in the Kew hari al oat triapiculata from specimens colle in Burma Robartaon, and o ro chin-China gathered bg Bien e Balanostreblus was described and figured by Kies in 1873. He quotes specimens under his B. ilicifolia from Chittagong (Hooker Thomson, No. and from Ava, Burma (J. An oe Through the kindness of Major Gage, it has been es e ine the material preserved in the Calcutta Herbari consists of the Chittagong specimen (of whic ere is more ple material at Kew), and o a plant marked as being “‘ culti- vated in the Botanic Garde It seems probable that this second specimen, which is fats was grown either from seed s Anderson’s at Kew were fro hamo, about 180 miles to the north-east. And as there is no eadcabt that this female cultivated specimen is the one figured and described by Kurz, it must be regarded as the type of the genus Balanostreblus. A sketch of ure. H Lu hegeteelt Chittagong specimen is male, and is undoubtedly able from 7. tlicifolia, Vidal, a ’ determination with which Gamble is is entirely in agreement. It is true that extreme forms But D. Merrill, oe in a letter to the Director first called attention to the similarity of Balanostreblus an rO- 2”? This variation is indicated in the text figures Mr. Meee soe come to the Se me - eed regarded as a form of B. ilicifolia, I can lee judge from the single speciait at Kew. The entire leaves of B. ilicifolia * Bureau in DC. Prodr. xvii, 217 (1873). 2% 148 (Fig. 5) are nearly always rather elongate oblong-lanceolate and not ovate-elliptic as in B. obtusa. i int that can best be settled in the field by Philippine botanists. : In the General Plantarum, Tazotrophis is included in the tribe Moreae, in which the anthers are inflexed in bud, whilst Balanos- CLAVIS SPECIERUM. Taxotrophis, Blume. Flores foeminei longe pedicellati ; — _ Flores masculi pedicellati : — 4 Qu oO B ee e Q | — iY) et ft) on aed —> ° be is") Dn +o nm °o — ar) ee iv} qi we » 1. javanica. 2. zeylanica. racemos i ms o Flores masculi sessiles; folia elongato-oyata vel oblongo-elliptica, caudato-acuminata 3. caudata. i os eemaaae of Taxotrophis; the numbers correspond with those in 149 Flores foeminei sessiles vel subsessiles : — Tnfloreavonitis 6 sessilis, crassa et densi- flora Folia apice Stee baie rotundata, in- tegra, elliptico-ovata . ... 4, obtusa. Folia acuminata, phon repando- -dentata vel integra et oblongo-elliptica vel ob- longo-lanceolat 5. tlicifolia. angerowenne o” pedun ne engust Ps ab ses ee vel laxiflor 7. Balansae. l= 7 a ih ae Fl. Ind Bat. i. ii. 279 (1859), eek syit Boerl. Ned. Ind. as 359 1900), Koord. et Val. Bijdr. path is xi. 4 (i 906) excl. syn.; Koord. Exkurs. -FI. Java. —Urtica ? spinosa, Blume, Bijdr. Fl. Ned. Ind. 507 (i picts spinosa, Blume, l.c Salak Mountain, Blwme; South-east Java, Forbes 1204; erent of East Je ava, up to 500 m. alt. (fide Koorders, l.c.), Vernacular. ‘‘ Panawar Beas’? (Sundanese). 2. T. zeylanica, Thwaites, Enum. 264 (1861) ; od 3 Ul Bot. iii. t. 11 (1851); Bedd. Fl. Sylv. . Sie 26, cores manicum ; ie Handb. Fil. Cerk iv. 100 ( par purus zeylan bc lneiaie: in Kew Journ. Bot. iv. 1. Diplocos ee as Bureau sheets Ceylon: fees near Kandy, Thwaites 2213. Trimen says this species is rather rare in low country from 1000-2000 og flowering in July, the male flowers yellow, the female gree Hooker’s saantihallion of Griffith’s Burma specimen (Kew w Dis- trib. 4659) was, if only from a et point of view, elongate-ovate, ca ate-acuminate leaves. fith’s a is flare’ made the type of a new species, Hutchinso . caudata, ee T. caudata, al sp. no 'T. zeylanica, Hoo £. Fl. Brit. Ss v. Pt. ii. 488, partim; Kurz, For. Fl. Brit. os ii. 464. 150 inconspicua, jae prominens, basi 1 mm. cra assa ; niervi laterales utrinsecus circiter 8, arcuati, ne prominuli ; petioli graciles, teretes, 3-5 mm. longi, puberuli; spinae axillares durae, usque a 15 cm. longae. Inflorescentia revis, puberula. Perianthu segmenta late ovata, obtusa. Antherae primum valde inflexae, demum erectae; filamenta gracilia, glabra. Ovariwm rudimen- tarium conicum, glabrum. Inflorescentia axillaris, 3-4-flora; pedicelli 5 mm. longi, in fructu —— ad 1-5 cm. longi, fere scentia, gla bra. Ovariwm valde obliquum ; stylus ‘crassus, ramis hispidis. Fructus oblique globosus, nitidus, circiter 6 tro. Ma. Upper pe ae pacha woods towards ite Serpe entine Mines, Apr. 1837, un Griffith (Kew Distrib.) 4659 ; without definite locality” 6, Hain Assam. Nowgong District: Lumding, shee 10, 1913, a small Kontal 9 tree 20-25 ft. high, with milky juice, g U pendranath an 4. T. obtusa, Himer, Leafl. Philipp. Bot. v. 1813 (1913). Puiirries. Palawan: Puerto Princesa (Mt. Pulfar), Apr. 1911, 6, Elmer 12966. 5. T. ilicifolia, Vidal, Revis, Pl. Vasc. Filip. 349 (1886). Balanostreblus ilicifolia, Kurz in Journ. As. Soc. Beng. xlii. ii. 247. partim, oe spec. Chittagongensem; et For. Fl. Brit. Burm. 465, partim; Hook. F. Fl. Brit. Ind. v. 544, partim. T. triapiculata, Gitihies in Kew Bull. 1913, 188. Inpra. Eastern Bengal : Pe Ges Hooker §& Thomson. Burma. Bhamo, Sept., /. Anderson. Southern Shan States: Kengtung; Mong Nai, alo ae streams in damp limestone gravel, 240 m., Mar. 9, 1911, Robertson 254-7 eee Proy. Bien Hoa: Mt. Lu, Mar. 1877, Pierre Matay Peyinsuta. Lower Siam : Trang Island, Mar. 1881 King’s Collec tor 1435. Perlis; Kanga, Mar. 1910, Ridley 14956; Kedah; Alasta, Feb. 1910, Ridley "14958 ; Pulau Adan ng, 2585 1911, Ridley 15714; Penang Waterfall, Apr. 1 1890, Curtas os Mar. 1915, Ridley ; Pahang; Tsmerloh, Ridley 2809a. ls ArcurPetaco. Celebes: Minahassa Province, Koorders Putt ar IsLanps. a ae localities, Vidal 904, 905, 1783 ; Loher 5222, 5223: . Marinduque Island, outh of Luzon, Vidal 1794, Ti95. "Pasian Island, Vidal 3770. 151 Basilan Island, W’. Z. carries For. Bur. no. 3443. Palaw Island; Puerto Princesa (Mt. Pulgar), Mar. 1911, Elaner 12858. Silanga, May 1918, Merrill 9603. In Luzon the vernacular name “ Cuyos cuyos.”” Mr. H Ridley informs me that the feves “of this plant in ii Peninsula are gilded over by the Tamils and used as a substitute for holly at Christmas; the wood is used for walking-sticks, and is so hard and heavy that it sinks in water. 6. T. laxiflora, Hutchinson, sp. nov.. Arbor parva dioica, frequenter spinosa; rami flexuosi, juniores laxe foliati, puberuli. Folia ilicina, obovate vel obovato- lon a, 23:5 cm. lata, acute spinoso-dentata, ere chartacea, sicco cinerea, glabra ; costa media supra angusta, prominula, infra prominens, straminea; nervi laterales utrinsecus 7-8, gracillimi, ilis, longus, hispidulus. Flores subsessiles albi. Perianthir seg- spe oblongo- —iiiioliek, acuta, fere glabra. Flores 9 ignoti. po-Curna. Tonkin: near Tu-phap, in the woods, Apr. 1888, Didone 2481, 2482. . T. Balansae, Hutchinson, sp. nov. is bor 7-8 m. alta; rami laxe foliati, sicco flavo-virides, glabri 1, profunde sule ati. Folia phish gecaltioticn vel obovato-elliptica, ong ities rotundata et leviter inaequilatera, apice in caudam 3 subacutam integrum abrupte vel subsensim nervis tertiariis et venis infra laxis; petioli 5-8 mm. longi, inter- dum verruculosi. Inflorescentia 6 spicata, ae gracilis, on 2 juniores non visi. ‘Perianthii fructifert segmenta " eoriaces, rotundata, usque ad 1 cm. longa. Fructus glaber, oblique obovoideus, ‘gts bilobato erecto coronatus; pedicelli usque ad 1 cm. longi. po-Cuina. Tonkin: forests of the Lankok ae Mar. 4, 1885, ae Balansa 2477; near the Thuang-Lam, May 2, 1888, g, Balansa 2478; in the woods at Pa-phip, 3 May 1887, Q, Balansa 2479. Species imperfecte cognita. T. macrophylla, Boerl. Hand. Fl. Ned. Ind. iii. 359 (1900) : pig Exkurs.-Fl. = ii. 87 (in obs.). —Streblus macrophyllus, Blume, Mus. Bot. Lugd.-Bat. ii. 80 (1856); ss Fl. Ind. Bat. i., pars. ii. 278 oy Diplocos macrophyl Bureau in DC. Prodr. xvii. 216 (1873) 152 Matay Arcurerraco: Celebes; near Likupang, coll. ? The only specimen of Tazotrophis at Kew from the Celebes is one (a maul leaf) “goltscted by Koorders, quoted under 7. tlict- folia, with which it is apparently identical. Balanostreblus, Kurz. Descr. emend.—F lores dioici, masculi ignoti; foeminei race- mosi; perianthium cum ovarii parte inferiore adnatum, medio pastas orificio protrudens, ramis 2 brevibus recurvatis crassis verruculosis. rupa (ut videtur) perianthio carnoso verrucoso inclusa, monosperma.—Arbor parva, sempervirens, lactescens (ie) folia alterna, grosse spinoso-dentata et nervosa. B. ilicifolia, Kurz in Journ. As. Soc. Beng. xlii. ii. 247, t. 19 (quoad description et iconum, excl. specim. e Chittagong.). Ramuli divaricati, glabri, cortice Cinereo verrucoso obtecti. Folia petiolata, oblonga vel elliptico-oblonga, acute acuminata, basi breviter cuneata, 310 em. 1 bap, 2-5-4 Bel pratt lata, 153 grosse spinoso-dentata, dentibus utrinsecus 3-9, rigide coriacea, glabra, supra nitida, infra leviter pallidiora et laxe reticulata; costa supra paullo imm a infra valde prominens et semi- teres, basi circiter 15 mm. crassa; nervi laterales, utrinsecus em ve productis, supra vix evidentes, infra valde sigan nervis tertiariis laxis infra prominulis; petioli 4-6 mm. longi, crassi, transverse verrucosi. Racemi 2 e ramis annotinis pen- duli, eirciter 3.cm. longi, pauciflori ; rhachis robustus, puberulus; bracteae minutae; pedicelli crassi, 2-5-3 mm. longi, hispiduli. Perianthium cikeabex 4 mm. longum, apice perforatum, medio constrictum et superne ovario liberum, extra verruculosum, haud lobatum; ovarium carnosum, glabrum; stylus 1 mm. longus, ramis 1 mm. longis arcte recurvatis verruculosis. Burma. Known only from a cultivated specimen in the Cal- eutta Herbarium, probably grown from a plant or seeds collected at Ava by J. Anderson. XII.—ROSA GLUTINOSA. R. A. Rorre. For some time it has been suspected that the Rose long culti- vated and recently figured as Rosa glutinosa (Willmott, Rosa, p. Sm., at all events as figured by the authors, and a comparison ‘of aa, with the aid of a specimen from the Sibthorpian Her- barium at Oxford, kindly lent by Prof. S. H. Vines, reveals an amount 3 ebatawon that it seems desirable to clear up as far as possi Bo. osa aida: Sibth. & Sm., was originally described in 1806 (Sibth. & Sm. FI. Graec. Prodr r. i, p. 348), and based on “ KX. Sphacioticis.” g published in Sibth. & € Sm. F 1. Graec. t. 482. In this later work we find the additional references, ‘‘ Lindl. Ros. 95,” rage 5 pumila alpina, pimpinellae exacte foliis sparsis, spinis incurvis, aquate purpurea, Cupan. Panphyt. ed. 1, v. 1. aid urning to aout. Ko. we find the farther synonym * R. rubi iginosa cretica, i. 93, 125, t. 50,” with the localities, ab. in There £?, also the note, fort (v. . herb. ; up rahi: I ‘ust to Sir James Sam No a considerable amount confusion, clude more than a single species. R. glutinosa was thus based upon — from two differen sources, first on the ce Rosa cretica mon SSE wii 154 to which the locality cited presumably belongs, and second on ure and the specimen from which it was drawn, speci doubt we will Sonic them first. Redoute’s and Cupani’s speci- mens are not concerned in the original publication. There is an original sheet of Rosa glutinosa preserved in Sib- thorp’s Herbarium at Oxford, named, but unlocalised, except the specimen marked ‘‘a’’ only differs in having traces of these aciculi. In the younger specimens there is generally a copious development of the glandular aciculi on the branchlets. There is also a sheet from Sibthorp in the Natural History Museum, South Kensington. It is labelled as follows: “R. spent F. Graec., J. L.; aff. villosa, n. 8. Rosa parnasst, 1 th.” To this Crépin has added: ‘‘ Veros. R. Heckeliana, Tratt., Crép.”” “J. L.’’ represents Lindley, whilst the specific name, arnasst, Sibth., though not originally published, was 2 now come to Tournefort’s original Cretan plant, which Temains uncertain, for the description is inadequate and no specimen is known. Tournefort’s specimens are at Paris, and ar. atrovirens, and R. glutinosa. Under resinosa, Stern. Spach Redoute’s figure, and R. sv. ehbateg de Bot. ii. (1813), p. 118, a Rose from the French Alps. at Kew. It is a fruitin figured by Sibthorp, and from its softly villous leaflets and petioles is apparently a luxuriant form of R. H eckeliana, Tratt., 155 with rather large leaflets, and an ellipsoid, very glandular fruit {in fully mature condition). The only other Cretan materials at Kew are in fl , and are typical R. Heckeliana, colle Dr. A. Baldacci, in 1893, at Hagia-Pneuma, one o summits ot the Asprovuna Mountains, south of Khania, a me pair, Tratennick (Bos. Monogr. i 3) mentions a R. cretica, Vest., as differing from Tournefort’ ‘lant and irae is described by Sternberg (Flora, ix. 1. pp. 74, 76) as R. resinosa. For this RR. resinosa, Re ichenbach suggests (FI. Germ. Excurs . p- 616) Cré the hybrid origin, Re FU ites > but Crépin “make no mention of such a cross. R. Keller (Ascherson & Grabn es Mittel.-Europ. FI. 67) sate of R. resinosa, Ste rab a variety of &. eine. Herrm.., while ee ne same time (p-. 106) citing R. rubiginosa var. cae Red., ynonymous with R. glutinosa, Sibth. & Sm. e R. cretica, “Wall. Ros. p. 144, is a complete oe vaddiae the forms already toaisteaall and several others, so that it throws no additional light on the ques- “Lastl ; we have Pa s figure, which is in the Linnean in his ie ‘(v. v. sp. herb. Smith).’’ It is inscribed ‘‘ Rosa pumila alpina Pimpinellae erage an (sie) sparsis a Aapoad in- pa aquate purpae, Cupani, : 61.” There is also icket: ‘‘ Rosa cretica montana ph " ibeoauniie glatinosis et Inst. Cor The l reference is to Tournefort’s Corollarium, and Dr. e Daydon Jackson informs us that the ticket is in the handwriting of Pietro Arduino (1728-1805), who sent many plants of Italian and gar- en origin to Linnaeus. It isa foun Raeracaag and agrees in , Ros. li. , as was sus- whilst the fruit is broadly ellipsoidal and twice as large as figured by Sibthorp. The Kew ies of this name was obtained 156 so that it would be idsrestit in R. glutinosa are the result a a Sores ee cannot Fd = desirable ne —MISCELLANEOUS NOTES. Major S. M. Toppin’s Bequest to the Royal Botanic Gar- jor dens, “ae ‘the will of the late Major Toppin, R.A., who, we deeply regret to record, was killed near Ypres on Se tember years’ residence on the North-Western Frontier of India and in accompanie ed by type-written descriptive notes and in some cases by coloured sketches of the — wers. om a Rae eane reference to the collection is made by Sir J. D. who, commenting on oben a in ‘Chitral ‘Kew Bull. 1911, p. 210), wrote n extensive, carefully ticketed, and beautifully pre- serv even Bac tas of Chitral plants has been formed b Lt. Toppin, R.A., who has kindly sent me the Balsams it con- : : t 157 sules in various species of Jmpatiens, indicating most methodical and painstaking work. In the mounting and arranging of the aa ¢ ; of his mother, who encouraged his early interest in botany, and to whose devotion and d_accomplishments his marked efficiency as a collector was in so e due Sidney Miles eae was fie younger son of Major-General J. M. Toppin, late of the Royal Irish Regiment, and Mrs. Toppin, as cease porege Branksome park, Bournemouth, and s born June 12, 1878, at Clonmel as educated Clifton College, and at Gonvi ie sh Gata Coleg, Cambridge, was working for a medical d ree when he was s offer eee ¢ in Hindustani he was given the charge of a “native Ch eens and went to the Afghan Frontier. While stationed at Chitral, where he spent two years, he applied his spare time to studying oud | plenty plants, Senenys. these pursuits later on when d to Northern Burma. in- ing his company he focver an appointment in Egypt. Pe cee ing to England in 1914 he married, and was about to leave for many by his untimely death. Major Toppin leaves a widow and infant daughter. His elder and only brother, Capt. H. S. Toppin, who did some excellent survey work (see Geogr. Journ., veh, mie p- 508), was killed in the battle of the Aisne, Sept. 14, 191 Presentation to the Herbarium.—A salloctibd of upwards 0 specimens of dried British plants made by the late Rifle- his father, Mr. J. J. Divers. John Divers was a gardener at Kew from March, 1912, to naa 1914, and was promoted to be Sub-foreman in the Herbaceous and Alpine Department in 1913. He joined the 25th London (Cyclists) Regiment in December, 1914, and later was transferred to the Queen Victoria Rifles. At the end of July, 1916, he went to France and after an engagement ‘on October 8-9, 1916, was posted as “‘ missing, believed killed. The collection received was commenced in 1908, when Mr. Divers was employed at Belvoir Castle Gardens, with plants from Belvoir and district, and was continued in the neighbourhood n im Reiga The specimens (entire plants where possible) have been aes with exceptional care and are well displayed on the sheets on which they are mounted. 158 but nevertheless appears to be quit good tanning material e aries in thickness up t somewhat smooth and of a pinkish-grey colour. It has been identified as Termi- nalia Arjuna, Bedd described by Gamble ‘AM of Indian Timbers ’’ as a large deciduous and beautiful tree found on the banks of rivers and streams throughout Central and Sou India, extending as far north as Ou Beyond that tow ards the North-west _ in the ue Be it is found only as a cultivated tree. It is also foun ur e low country o n Burma Ceylon. With regard oe ‘the applications of the bark in India it is employed both i in dyeing and tanning and in medicine as tonic and to heal wounds. In ‘“‘Pharmacographia Indica Dymock gives the following particulars of the chemical compo- ark : — sition of = :—‘* This is most remarkable, the ash amounts ih of almost pure calcium carbonate, which, if calculated Ht oxalate would amount to 43°5 per cent. The watery extract is 23 per cent. with 16 per cent. of tannin; very little colouring matter besides the ‘anti is extracted by aleohot. The tannin gave a blue-black ee ia ferric salts.”’ note on Myrobalans, the os n fruits of an allied wikia Terminalia Chebula. Retz., an important tanning oe appeared in Kew Bulletin, 1909, p. 209. “yee ag Bay Oil and Bay Rum.—The manufacture of bay oil and bay rum are important industries in the West Indies which have been developed chiefly during the last fifty years. Bay oil is the product of the leaves of the ‘‘ West Indian bay tree,” Pimenta acris, Kos Ly which is known locall by tay other names, and the oil is used in the manufaciure Sask is largely used as a hair wash. The leaves are etiared3 chiefly from wild sources, and the volatile oil re by a process of distillation The industry has suffered much from the eae of the leaves of two forms of Pimenta acris known ** Bois d’Inde Citronelle ” aid ae Bois d’Inde Anise,’’ which are so similar as to be separated with sieved € ** Bois d’ A Citronelle”’ is known of bay oil. The cand economic plant is known as ‘‘ Bois d’Inde,”’ or bay rum tree. i ogee of the leaves of these two varieties among aa produce sold ‘ Bois d’Inde ”’ leaves is not only a matter of ae she eaatahe and neg loss en us = Report on the Agricultural Department, Dominica, 1916-17, p. 2. 159 Leaves and flowering branches rs all three forms have been submitted to Kew by Mr. J. Jones, Curator, Botanic Garden, by which they can be didn guished, but the principal difference ¢ leaves, that of the : Bois d’Inde Citronelle ”’ being citron-scented, while the others laboratory, but no conspicuous differences have been observed. Mr. Jones remarks that the varieties have been cultivated in the Botanic Gardens for fifteen years, and that they were é cymes larger than those of Bois d’Inde Anise and several times larger than — produced by the true Bois d’Inde. Fru stated hee the eer have not Sa . fruit. Any siterbaoes ere, however, if they exist, would be of no practical value as a as in —- ‘the leaves store the fruits are develo ped. elie fi merce are mixtures of leaves of several different species In the event of the plant eae brought under cultivation, as has been suggested, owing to the inaccessibility or destruction that the plants selected had been derived from a pure stock of true Pimenta acris, Kostel. At present the leaves are collected from the wild plants in several islands, including Dominica, Porto oe Montserrat, St. John’s, the Virgin Islands and other localitie i he oil be interesting if, as suggested by Mr. Jones, accura observations were made on these points 1m all os Stands in which Pimenta acris is native, and in which bay oil is manu- factured, with a view to clearing up these ing atientene questions. This case of varietal forms of Pimenta acris affords a parallel to those of Camphor, Cinnamomum camphora, and Chicle Gum, — Sapota (See Repo ort on Agric. Dept. Dominica, 1914-15, p- 11) f both these plants more than one form i own exist and the present ference affords yet Guraliee example of the making sure of the value of any particular strain or physiological form of an soon plant before estab- e 160 Mr. Jones informs us that Pimenta acris and its varieties exist in great numbers on poor soils near the coast. e on bay oil and lemon-scented bay oil, and their distil- lation, with figures of the necessary stills, appeared in the West Indian Bulletin in 1908 (vol. ix. pp. 271-277), but no mention is there made of ‘‘ Bois d’Inde Anise.”’ When the lemon-scented Pimento was brought to the knowledge of Kew, some 85 years ago, the plant was propagated and dis- tributed between the years 1885 and 1889 to Jamaica, Demerara, Lagos, Queensland, Fiji, Ceylon, Java and Singapore. Peaches of New York.*—Kew is indebted to the ing, ? n peach cultivation as an industry, or as part of their ordinary avocation, will regard as of highest value. No less than 86 of the more popular and important varieties are illustrated by coloured plates, admirably reproduced. The association of the peach in legend as well as in name was for long with Persia, but that country, in spite of the testi- mony of ancient writers, is no longer regarded as the native country of Prunus persica. The tree has established and mining its real origin. The view has been upheld for some time Mr. Hedrick brings forward evidence which seems to prove con- clusively that this is true. It is interesting to learn that th peach has a greater commercial value in the United States than ‘all other stone fruits, i.e., plum, cherry; alm and apricot, put together e supported s. Darwin, Thos Andrew Knight and others of lesser note that the peach is 3 derivation nd is not endo b r. Hedrick is grown. * * The Peaches of New York: Report ii. of the N.Y. Agricultural Ombre: Station, 1916; J. B. Lyon Company, Printers, Albany, N.Y. [Crown Copyright Reserved. ROYAL BOTANIC GARDENS, KEW. BULLETIN OF XIV.—THE BRITISH SPECIES OF MELANCONIUM. W. B. Grove. The species of Melanconium, like many others of the commoner Coelomycetes,* have been much confused by authors and col- ectors. I and in the genus Melanconium this has been perpetuated and as it were stereotyped by a few unfortunate errors in Saccardo’s Sylloge. The following account of the British species will show that they can be arranged in three distinct sections :— § 1. Melanconium (sens. strict.), with smoky-brown or blackish spores, exuding as tendril-like masses. § 2. Lamproconium, with bright coloured spores (in this case, ue). when beaten e Tal a thin black layer closely adherent to the he bark around the orifice. : : In the third section, owing to the want of this mucilage, the spores on their escape become scattered over the matrix as an * See Kew Bulletin, 1917, p. 51. ; Sua! ; (5882.) Wt. 196—794. 1,125. 6/18. J.T. &S., Ltd. G. 14. Sch. 12. 162 irregular effused powdery stratum, which does not in any way adhere to the surface. This powderiness, as will be seen, hes led to serious misconceptions. § 1. MELANCONIUM as a bicolor, Ves, System, TOES, Dp. o2,, tabete, Beet; . Melanconium betiebiiumn: Schmidt §& Kunze, Exsice. no. 208, 1819. Perhaps the most far-reaching mistakes have been made in regard to the Melanconia on Birch. The differences between M. bicolor, Nees and M. oe Schmidt et Kunze, as stated The descriptions of the two species are ven by Link, in i Species Plantarum, 1825, ed. iv, vol. 6, pt. 2, front he placed them in ieee enera. The starting point of the agen poeay: must lie in MZ. bicolor. The essential points described by es (l.c.) are that the fungus occurred in Oak, and that + ees were ‘‘eyférmig.”’ He figures them as distinctly « ovoid, but on the whole not much longer than broad. Link (l.c. p 92) ‘does not make the matter chante: he describes the spores as rn ieniog oblongisque,”’ and t habitat as ‘‘ in ramis ejectis variarum arborum’’; but aa he adds the synonym ‘“‘ M. discolor, Schmidt et Kunze exs. n. 157” (published in 1817). An examination of this exsiccatum shows that the spores tend to be distinctly obovoid (rarely subglo- bose or oblong), not opaque, with one guttule, and measuring on the non 10-12 x 6-8». This agrees with the fungus called The other fungus, which seemed to Link to be nistage rede is called s him (Ic. p. 94) ee elevatum, Lin cites as ‘its type “‘ Melanconium betulinum, Schmid t et tiie: exsice. n. 208” (published in 1819), and gives the habitat as “ corticibus Betularum et Fagorum emortuis.’”? An examination obovoid and a pele Oe or even oblong, but es usually a very different ao. from those arieed by Nees. This is the fungus called below 0. 2. On investigation of the ; deahinleits issued oy. tp ‘mycologists 163 that M. betulinum was I-septate and M. bicolor not so: the answer lies in a peculiarity of the contents of the spores. In M. bicolor there is, in the majority of cases, a clear central oil- guttule, which ma owever, sometimes le towards either end, drop with a square outline towards the centre of the spore. In many, but by no means all of them, there is therefore, under a low power such as Link used, a distinct simulation of a median septum. This appearance vanishes altogether under a higher power, but has misled several observers, Greville, Vestergren, ete. As regards outward aspect, there seems to be a decided tendency in M. bicolor to burst out through a roundish opening, and in M. betulinum through an elongated lanceolate fissure. But on a wider examination of many specimens t this rule is seen to be ris ohh = so ee exceptions that it is ee | to name the two s s by external appearance only. fact they may not be really. ‘distinct, “bk as Tulasne thought ‘0% extreme forms of one species. Personally I incline more and more to this view of them as the end terms of a lengthy series, but in the biel of cases the spores appear as if perfectly distinct, and m thousands of species accepted at the “preset poe both in fungi and in other groups, rest upon no better foundatio The following exsiccata, among others, have bese examined : — No. 1. M. bicolor, Nees (spores 10-12 x 6-8 1). “*M. discolor,” Holl, Schm. et Kunz. no. (167! ““M. betulinum,”’ Desm. Cr ypt. Fr. no. Moug. et Nestl. Stirp. pee no. 670! a Westd. Herb. Crypt. Belg. no. 182! Ellis, N. Amer. Fung. “a 960! MM. "bicolor, > Wl. et Ev., N. Amer. Fung. ser. ii. no. 2390! (the same as no. 960, and sent out as a correction of the mistake). No. 2. M. pear Schmidt § Kunze (spores 13-16 x 5-6 y). “©M. betulinum,’? Schm. et Kunz. exs. no. 208! ‘“‘M. bicolor,’ Roum. Fung. Sel. ve Noor ”» ; Fekl. Pung. Rhen. no. 84! ” ; Syd. Mycoth. Germ. 143! Herb. Gerard, 2 Poughkaspete, se ve 33 Herb. Berk. ‘“‘ Thorney, Cambs.’ "Herb. Curtis, “ eee ee ae s levatum,”’ Tab rd. no p-p- Dilyncsspiere e oe: Parts e100 betulinu m, &S 4 ”? ae 164 — 6 60), which seems to his own showin ree Pp: 2a) on Birch and, as he ey states, is identical with M. Batis linum, Schmidt et Kun Saceardo wrongly — this to Dichomera Saubinetii. M. a Corda, therefore entirely disappears. The same may be said of Didymosporium profusum, Fr. Syst. Mye. iii. 487 = Stilbospora profusa, Grev. Scot. Crypt. Fl. tab. 212, f. 1 (1826). Original specimens in Herb. Kew, with the name written in Greville’s own hand, are on Birch and are identical with M. bicolor, Nees, t ongh he certainly included u he nam several of the other s s he supposed reidentification by the writer of Grevyille’s ae in Journ. Bot. 1886, p. 197, turns out to be an error; at that time the examining Greville’s specimens, and his figure is decidedly mis- leading. The peculiar shape and position of the single guttule gives under a low power, in spores taken from the original speci- mens mentioned above, a very deceptive simulation of a median septum he same e misconception no doubt accounts for some of the sith species placed under Didymosporium. 3. ——— rae a §& Ev. in Peck, 44th Rep. New York State Mus., 1890 This fungus, eo on ee virginica from Towa, U.S.A is Sunes Es in Herb. Kew by a specimen issued by Ellis, North > n Fungi no. 961! named ‘“ M. bicolor,’’ a name shoe - was ventiners ctitrocteid by the authors theniselve s, who the true M. bicolor in Sue = , no. 2390, and tien tend: ne name of no. to M. zonatum. Hit therto this species has eee been considered British, but there isa specimen in Herb. Berk. no. 1574 erpertatls British, ae no locality is given) on Birch, misnamed ‘ “on spores of this er from M. bespkor exact] bie assigned to M. zonatum, and it is no doubt “Soe oo Ostrya and Betula are Sine allied gener: The following is the description : the British specimen : — Pustules scattered, black, round, about 1 m x middle by a . semipellucid zone which in ees is seen to be caused by a large vacuolar space (not oil guttule : ) hark of Birch (Herb. Berk. no. 1574!), ade with Libertella etu The e-specimen on Ostrya differs only in having more dis- tinctly ae or subeylindrical spores, measuring 13-15 x 6-7 p35 the spores have the same dusky colour, whic ic different from that of 1. erties and the same median zone, due to the same cause. The whitish stroma is not really wanting, as Ellis ‘165 and Hverhart allege; a horizontal section of their own specimens will show it, but it oes not protrude through the spore-mass and make a white “eye”? so often as happens in M. peolms ae British specimens are A ge terme) intermediate between M. zon. tum and the typical M. bicolor, of which the org! might ai be considered to be only a strongly marked varie 4. Melanconium stromaticum, Corda, Ic. aes i. 3 (1837); Sace. Syll. iii. This species presents the usual source of confusion, in that it was considered by the seers authors to include what we should now call several speci is one certain guide avail- i seen ows ‘‘in ramulis Betulae albae.’’ What — was thinking = is iat that form of M. bicolor which doe as a matter of fact, occur, everyw here upon the smaller ‘ircucidets bicolor, var. ramulorum, to a Melanconium on at se being misled by what Fuckel had geeriouily done (Symb. Myc. p. 188). n examining the pub = sree one finds het speci- mens identical in all respects have been issued under the names M. stromaticum and M. = ies It will be sae therefore, to retain the former name for ce species on Carpinus, which is seen even with the naked eye to be different fren: aithars, and agrees fairly well with Corda’s “tigre in Sturm except that it The following exsiccata have been examined :— ee Fe ib Sts Fekl. Fung. Rhen. no. 90! s Rabenh. Fung. Eur. no. 1290! “