i i ‘ : ' ¥ : f < ‘ { ‘ - ‘ : , bs , ‘ ™~ 4 4° 3 / 4 P ‘ ‘ if ee i Ae as 7 theese ’ ; =k ae : 4 at - ; r 7 ‘ - Digitized by the Internet Archive in 2010 with funding from University of Toronto http://www.archive.org/details/ootamogetonspondOoitrye BRITISH POTAMOGETONS (POND WEEDS) So eee Ty =’. PLATE ] del,et hth : Vincent Brooks Day& Son Imp I, Reeve & C°London. "i THE POTAMOGETONS (POND WEEDS) OF THE Piet bebe Sick S Bee PESCHI TIONS OF ALL. THE SPECIES, VARIETIES AND HYBRIDS BY mERRED: FRYER, AES. AND ARTHUR BENNETT, A.LS. (The Lucens Group edited from MR. FRYER’S Manuscript Notes by A. H. Evans, F.Z.S.) ILLUSTRATED BY POBERT MORGAN, F.LS: AND OTHERS LONDON i KREBVE & CO: Liwitep 6, HENRIETTA STREET, COVENT GARDEN 1915 LONDON : PRINTED BY WILLIAM CLOWES AND SONS, LIMITED, DUKE STREET, STAMFORD STREET. S.E., ‘ AND GREAT WINDMILL STRHR?, W. eee ee OG) Na BENGE PLATES PAGE INTRODUCTION Vil KEY TO THE SPECIES xi GENERAL DESCRIPTION OF THE Guave Panera eran l British SpeciES AND Hysrips-— 1. natans 1-3 4 2. crassifolius 4-5 5 3. fluitans ; : : : é : : : : : 6-7 1 4. Kirkii . . , F : , s : F Z s . 8—9 16 5. polygonifolius ; : : ; ; : : . 10-12 18 6. coloratus : : : : 2 : 3 , , F =, What ls: 22 7. Billupsii : , : 5 : ; - : . 16-17 25 8. alpinus : ; : : : : : 18-20, 26 eas Gist, a 27 9. Drucei : : : P : : ; : 31 10. Grifhthii ; z : : ' : ? : : : ; a 34 11. praelongus : : : ; : : : : - . 24-26 36 12. perfoliatus 27-28 38 13. crispus 19-30 43 14. Cooperi 31-32 48 15. Bennettii 2 53 17. densus . F 34 56 The lucens group, ¢ edited by A. H. Evans . = 59 18. lanceolatus : p 39 63 19. graminifolius 36 64 20. heterophyllus 35 65 21. varians 41 67 22. falcatus 40 68 23. involutus 44. 69 24. nitens . 37-38 69 25. lucens. 45-47 i 26. Zizii (= angustifolius) 42 72 27. coriaceus 43 74 28. decipiens 48 75 29. salignus ; 49 76 Grass-leaved group, “by Arthur Bennett = 77 30. compressus (zosteraefolius) : 50 77 31. acutifolius 51 79 32. obtusifolius D2 80 30. Eriesii . 35 81 34. rutilus 54 82 35. pusillus 5D 83 36. Sturrockii (Fig. 1) 85 37. trichoides 56 85 38. pectinatus D7 86 39. vaginatus 58 87 40. interruptus 59 88 60 89 41. marinus 42, pennsylvanicus INDEX TO GROUPS, SPECIES AND acheauae PLATES, — = 5| 4 ~ fo) bo te] a INTRODUCTION BY ARTHUR BENNETT, A.LS. THE present work was projected some time ago by my friend the late Mr. Alfred Fryer, who had devoted many years to a most careful study of the British water plants included in the genus Potamogeton and their various states and forms, many of which he had kept under continuous observation by means of artificial cultivation. In the first few pages of the work he describes the general characters of the genus, and refers to the manner in which he proposed to deal with some of the problems connected with the systematic arrangement of the British species and natural hybrids. At the time of his death he had completed and passed for press the descriptions of the first seventeen species and hybrids, and had made considerable progress in the preparation of the matter for the next group, the Jucens group (pp. 59-76). Mr. A. H. Evans, who edited this section of the work after Mr. Fryer’s death, wrote as follows in regard to it :-—‘‘ With the exception of the distribution of the various forms, contributed, as in the earlier parts, by Mr. Arthur Bennett, the whole of this instalment may be taken as the result of the labours of Mr. Fryer on the section of Pondweeds to which he chiefly devoted himself. He left in MS. the introductory to this ‘/wcens-group, and he also sketched out the specific portion as a whole. A few of the articles were quite finished, and I am glad to say that I have found it possible to complete the remainder either from the author’s own ‘ Notes on Pondweeds’ in the ‘ Journal of Botany ’ for various years or from the information which he constantly gave me during the expeditions which we made together in the later part of his life to the localities near Chatteris, where the rarer plants grow or grew. I revised with Mr. Fryer the introductory paragraphs, and feel certain that the whole of the letterpress represents his final ideas on the group. If he had lived a few months longer he might have added further details, especially with regard to the land-forms, or what he considered mere states of the species he recognised; and therefore his former ‘Notes on Pondweeds’ should be read in conjunction with the present instalment, though it should be remembered that his mature views are not likely always to be in accord with those of his earlier years, when he was still engaged in struggling with the difficulties of his subject and first observing the growth of the plants in his tanks or in their native waters.” vill INTRODUCTION For the remainder of the work, dealing with the grass-leaved group, the present writer alone is responsible. I have had no help from my old friend’s MS. All that he left for this section were the names and synonyms of the species, and these I have adopted almost as he left them. Although Mr. Fryer had not taken quite the same keen interest in the grass-leaved Potamogetons as he had done in the other groups, it will always be a matter of regret that he did not live to work them out in the same exhaustive manner, and himself complete this monograph which will ever remain a monument to his memory. Mr. Fryer at one time suggested that the fertilisation of the genus was mainly brought about by small aquatic beetles, but though he was a keen entomologist, he does not seem to have followed this up, and I am not aware of any writer on our aquatic beetles who has suggested any such connection between them. Mr. F. Balfour Brown, in an article on the “ Aquatic Coleoptera of the Norfolk Broads,” * does not mention anything with reference to such a connection. In our own country the first to put the genus in anything like a systematic form was Sir J. E. Smith in his “English Flora,” 1824-28, where he described thirteen species, eight of the broad-leaved and five of the grass-leaved species. From that date to the year 1904, when the 9th edition of Babington’s Manual of British Botany appeared, the genus has gradually been added to and revised, until our Potamogetons have become fairly well known, though so far as their life-histories are concerned there is yet much to be learnt. The genus was instituted by Tournefort in his “‘Tnstitutiones rei Herbariae,” p. 232, 1700; it was taken up by Linnzus in the first edition of his ‘“ Species Plantarum,” p. 126, 1753, where he gives twelve species. Lourerio, in his “ Flora Cochinchinensis,” p. 301, 1790, uses the name “ Hydrogeton heterophyllus” for a species, perhaps one of the natans group. Rafinesque, in the “ Medical Reports,” New York, vol. v. p. 354 (1808), varies the name to “‘ Potamogiton ” » and Clairville, in his ‘‘ Manuel d’Her- borisation en Suisse et en Valais,” pp. 34, 44 (1811) has “ Potamogetum.” Rafinesque, in Journ. Phys. T. 89, p. 101 (1819), suggests “ Peltopsis” for P. perfolratus, L. Gay, in Compt. Rend. Acad. vol. xxxviil. p. 702 (1854), has “ Spirillus ” for the American species P. hybrides and spirillus, and on p. 703 suggests the name Groenlandia for P. densus, L. ; but none of these names have been taken up, as the genus is a very natural one, and does not need to be broken up. It is spread all over the five divisions of the world, but it is far more common in temperate than in tropical countries. In the Monograph of the Order in Engler’s “Das Pflanzenreich,”’ Heft 31, 1907, Dr. Graebner recognises eighty-seven species, with many sub-species, varieties and other forms, and about forty-seven hybrids; but * Trans. Norf. and Norwich Nat. Hist. Soc. vol. viii. pp. 58-82 (1905). INTRODUCTION ix these numbers are at present merely tentative, as the genus as a whole still requires much careful investigation and research before anything approaching finality can be affirmed. Since 1907 about eleven species and eighteen hybrids and sub-species have been described, with one omitted by Graebner in 1901, namely, P. Columbianus Sukesdorf. Forty-two of these species are limited to one country each, and using A. R. Wallace’s Areas of Zoological Distribution,* twenty-two species are confined to the Palaearctic Region, eighteen to the Nearctic, fifteen to the Neotropical, five to the Ethiopian, one to the Oriental and nine to the Australian Region. The remaining twenty-eight species are more or less distributed over all the regions, but one only, P. pectinatus, L., is found in all six regions. According to the Kew arrangement; two species are confined to Europe, six to North Africa and the Orient, three to Northern Asia, nine to China and Japan, two to Malaya, four to Australia, two to Polynesia, two to Tropical Africa, four to the Mascarene Islands, fifteen to North America, one to Central America, one to the West Indies, four to East Tropical South America, five to Temperate South America, while there are no species confined to India, New Zealand, South Africa, or West Tropical America; here again P. pectinatus, L., is recorded from all these divisions except Malaya. Of these the rarest are P. sibericus, Ar. Benn (once gathered), P. Drummondi, Bentham, gathered by Drummond, ¢. 1844, and not found since, P. Griffithii, Ar. Benn., P. Curtisii, Morang, in Florida, and three or four later named species which at present must be held tentative as to distribution, especially the Japanese. These islands are very rich in the genus, and have not yet been thoroughly worked, though there is a probability that two at present endemic North American species occur there. But this is quite in keeping with the character of the Flora, which the late Dr. Asa Gray has commented on. fF The limitation of the species is undoubtedly somewhat difficult, as some of them are exceedingly variable, being split up into petty varieties which are merely the description of individuals, whilst happily others will not lend themselves to be so divided. Considered from the local point of view this splitting up into varieties is of value so long as they are held to be local variations ; studied in connection with their environment as to colour, depth, density, chemical composition of the water, ete., they would in time evolve many interesting facts on their origin and evolution. To any one with a critical faculty and spare time to grow them, the genus would prove very attractive, as although Mr. Fryer grew many of the broad-leaved species, * A. R. Wallace, “ Island Life,” ed. p. 2, 31 (1892). + In Am. Jour. of Science, vol. 11, series 2, p. 135 (1846). Mem. Am. Acad. vol. vi. p. 377 (1859). b Be OF Cee SS * At eee " INTRODUCTION he did not grow the grass- -leaved species, and these would well repay the Asie is entailed, e eeah- They are not difficult to grow, but if grown in pots submerged, it is well to = cover the earth in which they are grown with a thin stratum of clay. Ifin tubs this = is not required, and while they will succeed in almost any soil, a mixture of one-third — ay Re of sharp sand is beneficial. In the last section, such species as prctinatus require more > room for rooting than the grass-leaved species, as they form many tuberous roots extend in every direction. If grown in a pond the margin is better than deeper oe except for such species as ea lucens, and perfoliatus which are strictly : submerged species of deep water, and never form floating leaves. KEY TO THE SPECIES OF POTAMOGETON. 1. Stipules free from the leaf. a. With floating leaves. Floating leaves various, coriaceous, Fruit large. Submerged leaves bladeless . - . : - - . natans, p. 4, pl. 1-3. “ » With blades é ; : ; - . fluitans, p. 11, pl. 6-7. Fruit small. Leaves all pellucid - : 3 3 : - - . coloratus, p. 22, pl. 13-15. 4) tins j : ; : ; . polygonifolius, p. 18, pl. 10-12. *5 elongated, 15 inches long : - “ - - . Kirkii, p. 16, pl. 8-9. Floating leaves ovate . : - ; - : - pennsylvanicus, p. 91, fig. 2. Floating leaves lanceolate. Leaves with distinct chain-areolation . - ‘ 5 . lanceolatus, p. 63, pl. 39. Submerged leaves all linear, others various. Leaves petiolate, submerged lanceolate, freely fruiting . . heterophyllus, p. 65, pl. 35. Be 3 > strap-shaped : A ; . graminifolius, p. 66, pl. 36. », Clasping, - n “ . falcatus, p. 68, pl. 40. sub-sterile or wholly so - + - . nitens, p. 69, pl. 37-38. Submerged leaves not linear, Fruit acuminate : ; : - : : : : . alpinus, p. 27, pl. 18-20, and 26 (figs. 6-8). » rounded . : : : ‘ : . 3 - . Zizii (= angustifolius), p. 72, pl. 42. b. Without floating leaves. Stems not compressed. Leaves cucullate atthe apex . : - : = é . prelongus, p. 36, pl. 24-26. », perfoliate, not cucullate : : : : - . perfoliatus, p. 38, pl. 27-28. » petiolate, acute or rounded . : ; : : . lucens, p. 71, pl. 45-47. », sessile, rounded . = A A - : : . decipiens, p. 75, pl. 48. Stems compressed. Leaves usually crisped and serrulate ‘ : . erispus, p. 43, pl. 29-30,“ Leaves all linear, submerged, fruit not toothed. 5 principal nerves = - “ - : : A . Zzosterifolius, p. 77, pl. 50. 5 ee ; : ss z 4 acutifolius, p. 79, pl. 51.; Stems compressed, branched, no secondary nerves. 5-7-nerved, leaves fasciculed - ‘ s . Friesii, p. 81, pl. 53. 3-nerved, not fasciculed, glands large translucent . - . obtusifolius, p. 80, pl. 52. 1- 3-nerved, not fasciculed, glands small, dull . ; 5 . pusillus, p. 83, pl. 53, Stems slightly compressed, usually simple. Leaves 3-5-nerved . : - . - - ; : . rutilus, p. 82, pl. 54. Stems filiform, branched. Leaves 3-5 nerved . ! “ ; ; : . Sturrockii, p. 85, fig. 1. » setaceous, fruit toothed - : ° : - . trichoides, p. 85, pl. 56. 2. Stipules sub-adnate to the leaf. With floating and submerged leaves . : - - : “ . Griffithii, p. 34, pl. 22-23. 3. Stipules adnate to the leaf. Fruit large, style lateral. Leaves 1-nerved, nut with faint lateral ridges . : : = . pectinatus, p. 86, pl. 57. 37 OOH 5h A ag HALODS aa F F : . interruptus, p. 88, pl. 59. Fruit small, style faint, nearly central. Leaf-sheaths much inflated . : . ‘ P = . . vaginatus, p. 87, pl. 53. not 5: . . - - : : ; . marinus, p. 89, pl. 60. 4, Stipules ‘absent. Leaves all opposite, submerged . 4 : ° : F . » densus, p. 56, pl. 34. Other forms, mostly intermediate between their supposed parents and variable, described in this work, but not included in the above Key :— crassifolius, p. 8, pl. 4-5. varians, p. 67, pl. 41. Billupsii, p. 25, pl. 16-17, involutus, p. 69, pl. 44, Drucei, p. 31, pl. 21. coriaceus, p. 74, pl. 43, Cooperi, p. 48, pl. 31-32, Bennettii, p. 53, pl. 33, salignus, p. 76, pl. 49, THE POTAMOGETONS OF THE BRITISH ISLES. POTAMOGETON, Linneus. Flowers perfect, sessile, on a terminal or axillary pedunculate spike; perianth segments 4, incurved, unguiculate; anthers 4, sessile, 2-celled, inserted at the base of the claws ; carpels 4, with shortly styled or subsessile stigmas, l-ovuled ; fruit a compressed, ovoid, or rounded drupelet, with the embryo incurved. Perennial aquatic herbs, inhabiting fresh, or rarely brackish or salt water; some species are amphibious when wholly or partially forsaken by the water in summer, then producing stunted growths fully exposed to the air; these are always infertile, but continue the life of the individual plant by thickened subterraneous stolons. Roorstock, in its simplest form, a stolon creeping in the soil and producing a stem at every second joint ; stolons either few or numerous, nearly simple, or with numerous secondary stolons; perennial or sometimes only annual in species which are reproduced by winter-buds formed on the fruiting branches of the plant. Stem usually terete, sometimes compressed, or flattened and leaf-like; mostly annual, rarely with the lower part surviving, and throwing out fresh branches a second year. Leaves sheathing at the base as in P. pectinatus, or narrowly strap-shaped and sessile, as in P. pusillus, or membranous, broad, and amplexicaul, as in P. perfoliatus, or short stalked, as in P. lucens, and submerged in all these species; or the lower membranous and submerged, and the upper stalked, coriaceous and floating, as in P. heterophyllus, or the lower reduced to a petiole and midrib, as in P. natans ; extremely variable in shape (even on the same plant) from linear to orbicular ; alternate, or opposite below the peduncles and branches; and all opposite in one species #. deisus. STIPULES adnate in the Pectinati; or free, and clasping the stem in the rest of the groups, in which, however, the lowest stipule on each stem or branch often carries an adnate leaf, as in P. coloratus, or several of the lower stipules, as in P. crispus ; herbaceous and persistent, or scarious, and soon decaying after their office of protecting the young shoots and leaves has been fulfilled. 6 POTAMOGETONS r erowth of the plant; the others are probably due to temporary local conditions. Certainly all such forms as were observed by me for a few years, either growing naturally or cultivated, ultimately reverted to the ordinary typical form figured in our first plate. I have been able to watch plants resembling all those figured except var. linearis, and in no instance have I failed to trace a complete reversion to the ty pe. P. rOLYGONIFOLIUS ¢c. LINEARIS, Syme, Ex. Club Report, p. 37 (1875); London Catalogue, ed. 8 (1881); Botanical Exchange Club Report (1888), p. 191. * P. natans, L. form,” Ex. Club Rep., p. 312 (1891). This early state of P. natans was mistaken by the late Dr. Boswell Syme for a form of P. pelygonifolius ; it differs in no respect from typical forms of P. natans which I have had under cultivation, except that in its Irish locality, ‘‘ Long Range, Killarney,”’ the force of the stream keeps it constantly submerged so as to prevent the formation of flowers, and to a great extent of the upper broad floating leaves. In consequence of this arrested development of the flower-spikes the phyllodes are more fully produced, and become more persistent, and the plant more branched. Mr. R. W. Sculley has liberally given me a very complete set of specimens of this form, by means of which it can be traced through all its stages until the typical jointed floating leaves are produced. Figure 5 in plate 2 shows the usual state of ‘ linearis.” In the Ex. Club Rept. p. 136 (1886), under the name of ‘‘ P. polygonifolius var. fluitans” (corrected by Mr. Bennett to ‘‘ pseudo-fluitans’’), a form of P. natans is noticed. The Rey. W. R. Linton gathered this in Connemara. This plant, as Mr. Bennett suggests, is P. natans, a form intermediate between the state named “ var. > lanceolatus,’” Fieber, and “ linearis.” This, probably growing in less rapid water, is enabled to produce small but abortive flower-spikes. ‘These slender spikes occasioned some doubt as to its being true natans, but I have seen almost as delicate ones produced under unfavourable conditions in stagnant water. « P, NATANS v. PROLIXUS, Koch,” London Catalog. ed. 8, No. 1497, b. (1886). ed. 9, No. 1580, b. (1895). Koch, Fl. Germ., vol. 11. p. 775 (1844). Prats 38. The original specimen is figured on our plate, it was gathered by Mr. T. F. Mott near Leicester, in 1881. The central and right-hand figures represent specimens from the same locality gathered by Mr. HK. F. Cooper in 1892; both gatherings were in May, and serve to illustrate the persistency of the local conditions which produced a form unvaried for many years. Here at last we seem to have reached a true, permanent variety; the leaves of the early and later gatherings are alike semipellucid and almost membranous in texture, and agree fairly well with Koch’s description, “‘ Folia submersa saepe numerosa basi apiceque attenuata pellucida, tamen non ita conspicue ut in P. fluitante.”’ From this description (assuming that Koch had not a form of P. fluitans before him) it 1s P. NATANS 7 evident that Koch mistook the not yet emerged upper leaves for strong submerged lower leaves. This state of P. natans is rarely to be met with later in the season. Mr. Cooper found a specimen in July with typical nafans, and I once found it in a densely crowded pond under the shade of trees, as late as July, but then all the growths that had reached the surface were of the usual form. In the Fenland drains it is common in spring, at which season the vegetation is frequently kept submerged by the artificial stream caused by drainage mills; as soon as the water is left at rest the leaves reach the surface and become coriaceous. Mechanical submersion, in one case by crowding, in the other by motion of the water, causes the attenuated shape and pellucid texture of the leaves. In a glass vase I once noticed a membranous floating leaf of P. nitens which was accidentally lifted on to the edge of the vase, and so partially exposed to the air; it became opaque and semi- coriaceous in a single day, resuming its transparent membranous texture after being replaced for about the same time in the water. This I repeated experimentally several times, always with the like result. Seeing, then, that the shape and texture of the leaves of this form are due to temporary conditions, and revert to the ordinary type of the species during the life of the individual plant, I think we many safely erase the “var. prolixus”’ from our lists. On referring to Koch’s description, the first part of it will be seen to apply to a somewhat different state of P. natans. ‘* B. prolixus, caules et petioli in rivulis valde elongantur et folia in rivulis strenua fluentibus saepe angustiora oblongo- lanceolata fiunt.” Specimens well agreeing with this part of Koch’s description were found by Mr. C. R. Billups in St. Helen’s Canal, Hulme, Lancashire, October 15th, 1886. These have phyllodes 2 ft. long, and petioles 23 in., with narrowly elliptical or spear-shaped lamina 2—25 in. long. This is the form described as prolizus of Koch in Dr. Morong’s Naid. N. America, p. 14. Visiting the Lancashire locality two or three years later, Mr. Billups found the water of less depth, and all the plants of natans had reverted to the usual form. When growing in shallow water, which becomes dried up in summer, land-forms are produced, these are reduced to a short-stemmed or nearly stemless tuft of broad coriaceous leaves, but these are preceded in all instances by a pair of the narrow phyllodes, always longer than the broad leaves; if watched from the first these phyllodes will always be seen (plate 2, fig. 3), but as they soon decay they are seldom noticed in the specimens one generally sees in collections. When the land- form is produced entirely in the open air it does not flower; all the so-called “ flowering land-forms ” of P. natans that I have seen evidently began their growth in water, and this rapidly wasting away they had subsequently assumed the land-form. P. natans has been recorded from all vice-counties except Dorset, South Essex, 4 POTAMOGETONS study, chiefly of the living plants; in the few instances in which I have not been able to make direct observations I have acknowledged my indebtedness to others. My friend, Mr. Arthur Bennett, has most generously given me his latest infor- mation as to the vice-comital distribution of the species through the British Isles, and such help as to priority of nomenclature as I demanded ; at the same time he is in no way responsible for the use, or mis-use, of the valuable aid he has so freely given. I may in some instances use the more generally accepted specific names instead of those which, though prior in date, have neither intelligible descriptions nor traditional usage to support them; thinking it perhaps better not too hastily to set aside well-known names in favour of those which their authors failed to make known by either clear descriptions, good illustrations, or by the distribution of specimens among their contemporaries. I will here merely add that when my friend has published his complete nomenclature of the genus [ will gladly be the first to accept his final decision, having perfect confidence in his unequalled knowledge of the subject, and his unrivalled power of specific discrimination in this difficult genus. 1. POTAMOGETON NATANS, JZinnceus, Species Piantarum (1753). P. natans, Smith, Eng. Flor. vol. i. p. 229. Bab. Man. ed. 7, p. 871.. Hooker, Stud. Flor. ed. 3, p. 431. Syme, Eng. Bot. vol. ix. p. 26, pl. 1899. Cham. Linn. vol. ii. p. 207,)No. 20,Tab."6,-fig. 20. ‘Hieber; Pot. Boehm, p. 22, Tab. 2, fig. 9. Fries Nov. Suec. p. 28. Reich. Icon. Germ. vol. vii. p. 26, pl. 50. Irmisch, Pot. p. 5, Tab. 1, £ 1-16. Boreau, Fl. Centr. Fr. ed. 3, vol. ii. p. 598. Gren. et Godr. Fl. Fr. Morong Naiad. p. 18, pl. 25. HE. F. Cooper, Leicestershire Potamogetons, p. 3. Fryer, Journ. Bot. vol. xxiv. p. 237. Lond. Cat. ed. 9, No. 1580. P. Raunkier, Danske Blom. p. 68, fig. 31. Ascheron et Graebner Synop. Mittel Europ. Flor. p. 303. (1.) Tiselius, Potamog. Suec. Hxsic. Nos. 1-4. Sprciric Cnaracter.—Leaves all coriaceous, floating ; upper jointed to the petiole by a narrow, flexible, discoloured part of the midrib extending a little below the lamina; lower reduced to a long narrow petiole and midrib (phyllode), exceeding the upper and ultimately floating. Description : Roorsrock stout, with terete stolons not contracted at the nodes, white with numerous slightly-raised reddish glandular spots subspirally arranged. Srem stout, terete, ascending from alternate internodes, sparingly branched, often - simple, the basal internode faintly glandular, the glands becoming immersed in the second and subsequent internodes. : Leaves all stalked coriaceous entire flat; upper with a flexible discoloured joint at the base of the lamina, many-ribbed, with the principal ribs prominent on the underside, often with a strong fold at the base, cordate, or abruptly or gradually P. NATANS 5 narrowed into the petiole ; lower reduced to a narrowly linear phyllode convex on the upper side for the lower half but slightly channelled towards the tip, much exceeding the upper leaves in length, ultimately floating, and then sometimes spathulate or clavate towards the apex ; these linear leaves are succeeded by others of an intermediate character which are expanded into ashort lanceolate or spear-shaped lamina (usually jointless) and so passing into the ordinary coriaceous elliptical to orbicular floating leaves, but soon decaying like the typical linear leaves. Sriputes very long, often exceeding the internodes, acuminate, scarious or herbaceous, with strong longitudinal fibres, back with two prominent ribs forming a slightly-grooved keel, soon decaying, but the fibrous part persistent. Prpuncuus slightly thicker than the stem, terete, slightly flattened towards the base andchannelled, narrowed towards the spike, shorter than the subtending leaves, 2-3 times longer than the spike. SprKk thick, cylindrical, densely flowered, 13-2 inches long. Fruit obovate, slightly compressed, usually rounded on the back, rarely with an acute keel when fresh; acutely keeled when dry, with faint lateral ridges; dorsal margin much rounded, upper slightly convex, beak short, facial. Conour shining dark green, or sometimes bright green, the young leaves streaked and tinted with reddish purple. Hasir very robust, stem 1-6 feet long, growing in dense masses, which often completely hide the surface of the water. The clusters of phyllodes are often produced late in the autumn, and grow slowly throughout the winter ; in spring, as the stems lengthen, the intermediate spear-like leaves begin to appear, together with the broad upper floating leaves, the latter being involute until they reach the surface, while the linear and spear leaves are flat throughout their growth. When the mass of vegetation reaches the surface the lower leaves soon decay. The permanent broad upper-leaves emerge from the water while still involute, and about half their length is exposed to the air for a few days, when, their growth being nearly completed, they expand, and, sinking back, float on the surface of the water, every part of which they are enabled to occupy by means of their flexible joint, which allows each leaf to twist in any direction, even reversely parallel to its own petiole. The involute emerged leaves are conspicuous from a considerable distance, and serve to distinguish P. natans when growing from its associated species. Several named varieties of P. nataus have been described by authors ; of these it will be necessary to consider only those which have at some time found a place in the writings and records of British botanists. After a careful study of these forms, and the cultivation of most of them, I have come to the conclusion that, so far as this country is concerned, all these so-called varieties are mere states, some of which are merely the natural changes of form, which take place according to the progressive 2 E IT OTAMOGETONS FLower-sPik& usually emerged until fertilization is accomplis! ed, then it falls back, and the fruit is ripened at or beneath the surface of the water; very rarely fertilization has been observed to occur in wholly submerged spikes. FLowrrs proterogynous, as far as I have been able to observe, and probably fertilized by the pollen being discharged with some force from a contiguous spike, or by the action of insects, or by the wind, or even by being carried by the water itself. In the latter case the occasional fertility of constantly submerged flowers is easily explained. The Frurr affords good characters to aid in the discrimination of species, but it is difficult to observe, especially when fresh. When dried the external sculpture becomes more pronounced ; care, however, must be taken to found comparisons ou perfectly mature examples. Much importance is attached by some botanists to the exact degree of curvature of the embryo, and the relative thickness of the radicular and cotyledonary ends, but these features are variable, as well as the external form ; and I believe this slight degree of variation is not solely due to age or degree of maturity, but is often inherent, like the variability of the plants themselves. Hence the distinctions afforded by the fruit in closely allied or “ critical” species are of little practical use to the average observer. If a comparison is made of the figures and descriptions of the fruits of Potamogetons given by Chamisso, Fieber, Reichenbach, and Morong, the difficulty I have pointed out will become very evident. The drupelets are ovoid or subglobose, somewhat compressed, keeled, or rounded on the lower margin, with two fainter lateral keels; the nutlet is 3-keeled ; the embryo incurved, uncinate or cochleate, and always with the cotyledonary end somewhat ascending, so as to form part of a spiral; radicular end thickened. Although the species differ so widely in external appearance, the genus does not readily afford distinct groups, nor subgenera; the arrangement usually followed is that of Koch, who has divided the species into the following groups :— 1. HetERoPHYLLI; comprising the broad-leaved species with coriaceous floating leaves, from P. natans to P. nitens. 2. HoMopnyLti; species with all the leaves membranous and submerged; from P. lucens to P. crispus. 3. CHLOEPHYLLI ; species with strap-shaped, grass-like leaves, from P. Friesii to P, trichoides. 4, COLEOPHYLLI; species with adnate stipules, the bases of the leaves sheathing the stems. : 5. HNANTIOPHYLLI; leaves all opposite, P. densus. Looking at the Potamogetons of Europe these are fairly distinct groups, with the exception of one aberrant species, P. crispus, which is in many respects inter- POTAMOGETONS 3 mediate between the Homophylli and Chloephylli; but if we try to group the Potamogetons of North America in the same manner we fin] in P. spivillus a species which unites in itself the characters of Groups 1,3 and 4. Even in Europe the Groups 1 and 2 have some intermediate forms, as P. alpinus for instance, which is sometimes without coriaceous floating leaves. In other instances, as P. decipiens with coriaceous floating leaves, the intermediate forms are hybrids between species belonging to different groups, and so naturally possess characters derived from each. These points will be considered more fully when the species are described in detail. In the following pages those characters which either separately or in conjunc- tion seem peculiar to each species are grouped together to forma specific character, which is placed before the more detailed description. The student, it is hoped, may be able, by means of these short descriptions and a reference to the plates, to name the ordinary states of any of our British species. The detailed descriptions are drawn up in their present form for the purpose of affording a means of readily making a critical comparison of any one part of the plant throughout the whole series of species without the necessity of reading the whole of the descriptions. For the more or less permanently differing variations of those species which are capable of producing fertile seed the term variety is used ; for temporary differences (often more striking than permanent variations) depending on age of the individual or unusual local conditions, and which naturally during growth, or on the plant being restored to more natural surroundings, revert at once to the typical form, the term state is applied. Among the older and generally accepted species we find some, such as P. witens Weber (1787), and P. fluitans, Roth (1788), which have recently been proved to be hybrids. To reject these names would lead to great confusion, and would tend greatly to obscure our researches into the origin of many specific or quasi-specific forms ; in retaining them I also think it well to name newly-discovered hybrids, prefixing a X to the generic name (X P. fluitans) to distinguish these forms from true species. The first described form of such hybrid species is taken as the type; the departures from which are of two kinds :—when an individual plant of one of these hybrids has been observed to sport into a permanently distinct form, the term variety may still be applied ; but when a distinct result is obtained by the interbreeding of the same parents, or in cases where the parents are reasonably supposed to be the same, the difference is expressed in the following pages by the Latin word forma. No attempt is made at unravelling the intricacies of synonymy; the references to other writers are given solely to direct the student to the best sources of informa- tion. The specific descriptions and life-history of each species are made from my own 8 POTAMOGETONS + Denbigh, Flint, West Lancaster, Durham, Isle of Man, Kirkcudbright, and Banff ; probably it occurs in all or most of them. It is frequent in Ireland, common throughout Europe and North America, and has been found in several localities in Asia and Africa; although reported from Australia and New Zealand, it is probably absent from Australasia. Explanation of the Plates. PoramoGrron Natans. Plates 1—8. Praty 1, fig. 1, fruiting spike; 2, drupelet mag.; 8, longitudinal section; _ 5, section of the stem. 4, transverse section ; Prare 2, fig. 1, barren summer shoot, to show phyllodes and underside of somes leaf; fig. 2, land-form, early state with phyllodes. (All the above are drawn from Cambridgeshire specimens) ; fig. 3, “ var. linearis,” from specimens gathered by Mr. R. W. Sculley at Long Range, Killarney, “from slow-flowing water,” selected to show the branched stem with persistent phyllodes and the rarely-formed upper leaves. Specimens from “ swift-flowing water ”’ do not show these leaves. Prare 3. The lower figure on the right is from Mr. T. F. Mott’s original specimen gathered by him, ‘‘ May, 1881, pond near Scraptoft Long Spinny’”’; the central and left-hand figures are from the same Leicestershire locality, gathered by Mr. E. F.Cooper, ‘‘ May 5th, 1892.” 2. X POTAMOGETON CRASSIFOLIUS, Fryer, Journal of Botany, vol. xxviii. p. 321, Tab. 299 (November, 1890). P. natans X Zizi, Aschers. et Graeb. Synop. p. 332 (1897). SPECIFIC CHARACTER: a barren hybrid between P. natans ¢ and forms of P. coriaceus $ (P. Zizti, Auct.). Leaves all stalked, coriaceous when mature ; lower leaves fleshy, soon decaying, distant, few in number, upper oblong-oval or obovate, often strongly auricled at the base of the lamina, not completely floating, but often partially submerged like those of P. coriaceus, phyllodes spathulate, shorter than or little exceeding the internodes; stipules acute, not winged nor channelled on the back, whitish, with two conspicuous green ribs, horny with scarious edges, or green and herbaceous, the lowest rarely with a short adnate phyllode, or a short ordinary lanceolate lower leaf. Desorption: Roorstock stout, white with reddish glands, stolons with short internodes, scarcely contracted at the joints, deeply rooting. Svems stout, terete, 1-5 feet long, simple below, shortly branched above, or simple throughout, irregularly ascending at intervals of several internodes, frequently terminal. Leaves all stalked, coriaceous, opaque when mature; lower fleshy and trans- lucent when young, lanceolate, gradually narrowed into the petiole, lowest reduced to linear or spathulate phyllodes shorter than or little exceeding the internodes, all P. CRASSIFOLIUS 9 soon decaying, upper leaves oval, or oblong-oval, rarely elliptical, usually strongly auricled towards the base, with the lamina decurrent below the auricles; lateral ribs translucent, not sunk in the substance of the leaf, nor usually prominent beneath. The lowest leaf of each branchlet is often lanceolate or oblanceolate. All the leaves are thick and fleshy when fresh, but extremely thin when dried, thus resembling those of P. coriaceus. Pxtiote shorter than the lamina, flat above, expanded towards the leaf, herbaceous and veined with transverse veins throughout, thin when dried, like the leaves. STIPULES acute, horny with scarious edges, white with two conspicuous green ribs, but not winged nor channelled on the back, or green and herbaceous, 14-2 in. long; lowest sometimes bearing a short phyllode, or more rarely a somewhat longer ordinary lower leaf. PEDUNCLE equal, terminal or becoming lateral by continued growth of the branch ; equalling or slightly exceeding the stem in thickness, 2-3 times longer than the flower-spike, much shorter than the subtending foliage. Sprke cylindrical, dense, 1-1} in. long, barren, sepals remaining closed, but stigmas exserted, soon decaying. Fruit abortive, very rarely formed, but never perfected, whole spike soon decaying. Cotour bright green, the young upper leaves often streaked with reddish- purple, like those of P. natans. Hasit: lower leaves permanently submerged, usually decayed by the time of flowering ; upper leaves sometimes raised above the water till fully expanded, like those of P. natans. The plant often grows throughout the winter, like P. natans, which it resembles in its general habit. In the Journal of Botany I have described two varieties of this hybrid species (l.c. p. 8323), one of which I have had under cultivation for eight years. It never wholly dies down like P. fluitans, but the submerged leaves and stems grow slowly all through the winter. Although it produces flower-spikes in abundance, these rarely produce even abortive drupelets, but decay soon after the stigmas are exserted. In this form the upper leaves are regularly spread out and flat, orbicular-ovate, rarely lanceolate, with prominent ribs and with the purple colouring beneath of P. natans, and the whole plant is a little stouter in substance when dried than in the typical form. The “ Westmoor variety” I have now decided to name as a distinct form, possibly of somewhat distinct parentage :— x P. ORASSIFOLIUS, FORMA VERRUTA mihi. Plate 5. fo) 10 POTAMOGETONS Hybrid, P. natans 9 xX P. Zizi 3. Stem branched from the base or simple, submerged leaves oblanceolate or lanceolate, very long, sometimes exceeding the floating coriaceous leaves, phyllodes sometimes produced along the stem nearly up to the floating leaves, and then exceeding them in length; floating leaves lanceolate, oval, or oblong-oval, with the petiole often much longer than the lamina, which occasionally bends back as in P. natans, but without a perceptible joint, rarely auricled at the base, extremely variable in shape, and length of petiole; peduncle stouter than the stem, greatly exceeding the very short spike. Habit of P. natans, or of the flowering branches of P. Kirkii. A comparison of the figure of this form on plate 5 and that of P. Airkit on plate 8 will show the necessity of studying the early states of Potamogetons as well as the more perfect or flowering states. Much of the lower part of f. verruta might well pass for P. Kirkit. The young state, however, is totally different, being nearer that of typical crassifolius; while the occasionally elongate phyllodes and narrowly lanceolate lower leaves exceeding the short- petioled upper leaves, in some of the plants, approach the habit of P. natans. The plant figured is from a cultivated root, which was gathered in a shallow ditch at Westmoor, Doddington, Cambridgeshire; it was one of many seedlings which I was able to watch for several seasons. A full account of the circumstances under which they grow will be found in the Journal of Botany, November, 1890, © p. 322. Since that was written a single plant was found in a flowering state; it is so totally different in facies that I regret the limits of our work forbid a separate plate. If I had not been able to connect the flowerless and flowering form by numerous intermediate forms, I could not have even guessed that they were of the same origin, and many of them probably from the same cross-fertilized spike. In naming new hybrid-species in this work I am guided solely by their parentage, not by likeness to one type in themselves, nor by difference from already named forms. Else the forms of verruta might have had half a dozen names, as they are so dissimilar in appearance, resembling Kirkii, natans, Zizi, and with additional special forms of their own. The hybrids of Potamogetons are not like the hybrids of less variable plants ; in some instances they may be constant enough to one type, but in others the variation seems to have no bounds. We find suggestions of not two, but many species, and it is impossible to suppose that these resemblances are accidental or uncontrolled by law. When by observation and actual experiment the law that governs the variation of cross-bred forms of this genus is discovered, wholly or in part, we shall probably get a clearer view of one of the modes by which ‘‘ species” may have originated. The plants in this group of forms allied to P. flwitans are infertile hybrids, later on we shall have to consider fertile hybrids. In what respect do they differ from “true species’ P P. FLUITANS 11 Distripurion: P. crassifolius is at present only known from three separate localities near Chatteris, Cambridgeshire. Through the kindness of Mr, J. Baagoe, I have received Danish specimens which suggest a similar parentage, but owing to the extreme variability of some of the hybrids of the flwitans-growp I dare not name them. All such forms can be determined only by exhaustive observation in the field ; comparison of specimens is here useless, or even misleading. No doubt other different-looking hybrids between Natans and Zizii will be found, but from what has been observed in other extremely variable species there is little doubt that some common character will be found by which they may be separated from P. fluitans as clearly as they now are from P. Kirkii. When Natans grows intermixed with both lucens and Zizi it will be necessary to study hybrid seedlings that may be found, from their earliest stages. Explanation of the Plates. x PoramocETon orassirotius, Plates 4 and 5. Prats 4. Rootstock, phyllodes, lower and upper leaves; fig. 1, lower stipule with adnate leaf; 2, section of stem, mag. All from Mepal Common, Cambridge- shire, July, 1897. Prats 5. forma verruta, mihi; a complete barren branch from root upwards ; cultivated plant from Westmoor, Doddington, Cambridgeshire, August 8th, 1892. Peduncle and flower-spike, wild, from same locality, September, 1890. 3. x POTAMOGETON FLUITANS, Roth. Tentamen Flore Germanic, vol. i. p. 72 (1782), vol. 11. p. 202. P. riuitans, Cham. et Schlecht. Linnea, vol. ii. p. 219, No. 21 (1827). Fieber, Pot. Boehm, p. 21 (1888). Fries Novit. Suec. ed. 2, p. 29. Koch, Synop. Fl. Germ. ed. 2, p. 776. Reich. Icon. Germ., vol. vii. p. 26, Tab. 49 (excl. B stagnatilis). Fryer, Journal Bot., vol. xxvi. p. 273 (1888). Beeby, Journal Bot., July, 1890. Raunkizr Dansk. Blomster, p. 97 (1896). Tiselius, Potamog. Suec. Exsic. No. 6. Sprcrric CHARAcTER.—Stem branched from the base, lower leaves membranous, lanceolate, persistent ; upper oval-lanceolate coriaceous, all long-stalked ; stipules horny, herbaceous, persistent; flower-spike barren ; a hybrid between P. xatans ¢ and P. lucens &. Description: Roorstock with crooked stolons, internodes somewhat thickened and sub-tuberous, striking deeply into the soil. Stem moderately stout, terete, ascending at irregular intervals, much branched from the base, or simple in late growths; lower branches submerged, not flowering, upper branches floating and spreading on the surface of the water. Leaves all stalked, lowest reduced to narrowly linear, short phyllodes per- manently submerged, followed by linear-lanceolate, or narrowly lanceolate, or elliptical-lanceolate membranous translucent submerged leaves, somewhat undulated, o 2 12 POTAMOGETONS flat or longitudinally folded, acuminate, with the lamina insensibly narrowed into the petiole ; upper leaves coriaceous, floating, oblong-lanceolate, or elliptical-lanceolate, with the numerous lateral ribs usually translucent and immersed in the substance of the leaf on the underside, or with 2—3 on each side prominent towards the base of the lamina ; sometimes slightly folded towards the base, but these with a part of the lamina decurrent below the fold. PrrroLe slightly convex above, always shorter than the lamina. Strputys very long, 1—4 in., deeply channelled on the back, or sometimes slightly winged, acute, persistent, at first clasping, afterwards patent; horny in texture, often herbaceous, green or whitish with green veins; rarely with a leaf reduced to a phyllode on the back. Pupuncte 3-4 in. long, a little thicker than the stem, equal, or slightly swollen in the middle. Spike cylindrical, 1-13 in. long, densely flowered ; sepals rarely expanding, and then bright green, or closed and dull green, but the stigmas always exserted. Fruir abortive, 3 keeled, the central keel acute. Hasir: The whole plant rising by the time of flowering, and spreading in fan- like masses on the surface of the water, the leaves never reflexed nor irregularly disposed as in P. natans. By this time most of the translucent membranous lower | leaves become coriaceous like the true floating leaves which support the flower-spike, only a few of the lower leaves remaining submerged and membranous. In autumn iong narrowly linear folded grass-like leaves are often thrown out from the lower part of the stem, especially when the plant has been cut over repeatedly in the summer ; but these are quite distinct from the phyllodes which are constantly pro- duced on the young stems in spring. These submerged phyllodes gradually pass into the ordinary membranous lower leaves by intermediate bodkin-pointed leaves with the petiole and midrib thick and fleshy, but expanded in the middle into a narrow wing-like lamina as in P. lucens. Cotour bright green to olive-green, drying darker. This plant was first separated from P. natans by Roth in 1788, and described in his Beytrdge zur Botanik, but not named until 1788 in his Tentamen Flore Germanice. In reading his excellent description of the species it is necessary to remember that P. polygonifolius was at that time still confused with P. natans; bearing this in mind Roth’s description is so exact and clear that it is not a little remarkable how much error and confusion has gathered around such a well-defined form. Mr. W. H. Beeby in an excellent paper in the Journal of Botany (July, 1890) made the first decisive step towards clearing up the difficulty by showing that the barren hybrid form figured by Reichenbach (Icon. Flor. Germ. Tab. 49) exactly P. FLUITANS 13 agrees with Roth’s description ; and he also points out the difference between the specimens seen by him of the fruit-bearing European plant (often named “* flitans ’’) and the plant figured by Reichenbach :—‘ My contention, therefore, is: First, that the demand for fruit, wherewith to prove that any plant is P. fluitans Roth, must be withdrawn, or the evidence whereon the demand is based be divulged; and secondly, that the name P. fwitans Roth must be withheld from the freely-fruiting Continental plant, untilit is shown to produce the characteristic submerged leaves described by Roth.” . 2 Subsequently Dr. Tiselius generously lent me his fine series of the fruiting plant collected by himself in the River Neckar, and also gave me a very perfect set for my own herbarium. By these means I was enabled to compare the two forms, in both early and mature stages of growth, and I at once saw that the two plants setting the question of fruit altogether aside, were distinct in foliage and habit. The fruiting plant from the River Neckar* has, indeed, a few translucent submerged leaves, but they are more distinctly stalked than in the plant of Roth, and are all decayed by the time of flowering. No linear-lanceolate leaves nor phyllodes are present. The upper leaves are oval or oblong-oval, with the petiole often greatly exceeding the lamina, which has the lateral ribs raised above the surface on the underside, not immersed in the substance of the leaf as in P. fluitans. The lateral ribs are opaque even in young examples, instead of translucent, and the areoiation is totally different. The rootstock is more like that of P. polygonifolius, as also are the stipules, and the stem is simple, or rarely with lateral branchlets; finally, the habit of the fruiting plant is more diffuse, the foliage is more scanty, and the floating leaves do not spread like a fan. As far as I am capable of judging, the European fruiting plant is not specifi- cally identical with P. Lonchites of Tuckerman, neither in foliage nor in fruit; in the latter both external sculpture and curvature of the embyro seem different to my eyes. Before leaving the subject I would call attention to Mr. Bennett’s Notes ‘on Potamogetons, Journal of Botany, October, 1893; he there says :—‘‘ We have no certain knowledge of any of Roth’s species being preserved in any herbarium ; but there are at Munich specimens in Schreber’s herbarium, namea as such, and gathered in ‘In Seebach, 1775,’ and others ‘ In Seebach, 1782.’ It seems to mea reasonable inference that these specimens are from (or seen by) Roth; the more so * Afterwards considered by Dr. Morong to be equal to P. Lonchites, Tuckerman, var. Noveboracensie Morong, Naiadaceze N. America, p. 21. 14 POTAMOGETONS because there are other species in the same collection actually received from Roth, and signed by him. They are the plant we call fwitans in England (hybrid ?), and not the Neckar plant of Schimper and Dr. Tiselius.” I will now try to solve Mr. Bennett’s query as to the hybridity of Roth’s P. fluitans, especially as it will show the nature of the evidence on which I have assumed certain plants described in the following pages to be natural hybrids. This one case clearly stated will save much needless repetition. P. fluitans is found, according to Mr. Bennett, in “ Europe generally”; although having so wide a range, it has never been found in a fruiting state, although it has been repeatedly gathered and observed for over a hundred years ; this alone affords strong presumptive, but not positive, evidence of hybrid origin ; but if we add to this strong presumptive the more positive evidence afforded by its local distribution and its sudden appearance in new localities in the fens of Cam- bridgeshire and Huntingdonshire, I think no one can doubt its origin from P. natans crossing with P. lucens. When I first found it in 1884 it was growing in an isolated pit only a few yards square with the two supposed parents and P. crispus, but as the station is a low, swampy excavation by the side of Vermuyden’s Drain, and is lable to be submerged by floods from the upper or “ high-lands,” I naturally supposed that at some time or other the species had been carried thither from, or by means of, the brook which brings down the upland floods. A repeated search year after year clearly showed that both brook and river above the station of fluitans were not tenanted by the species. In the course of two or three years I found other stations where fluitans occurred, evidently from seeds, in single plants in ditches and drains which were far removed from, and had no connection with, one another. The next step was to try if this form, which had not been found seeding in a wild state, could be induced to do so under cultivation ; it grew freely, but the flower-spikes remained barren, and I began to feel sure that fluitans was a hybrid. All doubt was removed when I found a plant growing in a drain which had been recently deepened, and most of the vegetation destroyed, in a spot which had been constantly examined by me for years. This plant when first found was evidently of recent origin, and then confined to a single rootstock, and it was produced under circumstances which I have repeatedly found favourable to the intercrossing of distinct species of Potamogeton ; that is, by the drain having the greater part of its vegetation destroyed, so that single flower-spikes grew apart from others of their own species. As a rule, the stigmas are exserted and receptive before the pollen on the same spike is ripe; if no other spike of its own kind is near to supply the necessary pollen, then the stigmas may become fertilized by a contiguous polliniferous spike of some other species. At one time I thought this could only take place by mechanical, and so accidental, means, P. FLUITANS 15 but more recent observations show that the ripe pollen is discharged in a shower some little distance around the spike; thrown around it, in fact, with sufficient force to reach many or all of the stigmas in a contiguous spike. Usually only one or two druplets are thus made fertile, but the occurrence of hybrid seedlings along one ditch in the before-mentioned case of P. crassifolius induced me to believe that occasionally the whole or a greater part of a spike was fertilized by the pollen of another species. P. fluitans never occurs in the fens apart from P. natans and P. lucens ; in like manner P. crassifolius never occurs apart from natans and Zizii ; and generally when a drain or ditch has been cleaned out and deepened, so that the vegetation has been for the greater part destroyed, I have found after a year or two seedling, or appa- rently seedling, plants of hybrid Potamogetons. Short of producing these forms by artificial crossing, it seems impossible to obtain stronger evidence of their hybrid origm. When we further consider how exactly P. fluitans is intermediate in many of its characters between P. natans and P. lucens, as was long ago suggested by Chamisso and Schlechtendal and by Fries, it seems impossible to doubt its being a true hybrid, constantly reproduced from the crossing of these two species. At present P. flwitans is only certainly known as a British plant from the counties of Sussex and Surrey, where it has been found by Mr. Beeby, and from Huntingdon and Cambridge. Mr. Druce sent me specimens from Berkshire, which probably will be found to belong to the fluitans-group, but they are not sufficient to name with certainty. Out of England this hybrid species is at present known only from ‘‘ Europe generally.” See Mr. Bennett’s valuable note in the Journal of Botany for October, 1893, and vol. xxxii. p. 372 (1895). Description of the Plates, 6-7. P. riurrans Foth. Prate 6. Upper part of flowering branch of P. jluitans, from a specimen gathered in Huntingdonshire, September, 1893. To get so large a portion in the plate it was necessary to disarrange the fan-like disposition of the upper leaves; all the leaves were strongly undulated on the margin, some of the lower ones so much so as. to look serrulate ; in this Mr. Morgan has exactly reproduced their natural appearance in the fresh specimen! This undulation disappears to a great extent when the plant is dried under pressure; all these leaves had entire margins, although some of them were almost crisped when fresh. Prate 7, fig. 1. Part of the rootstock showing young growths with phyllodes and lower membranous leaves; fig. 2, the Land-form, with limear lower leaves ; fio. 3, axillary branchlets by means of which the plant is propagated in autumn. 16 POTAMOGETONS r 4. x POTAMOGETON KIRKII, Syme, Eng. Bot. ed. 3, vol. ix. p. 31 (1869). Hooker, Stud. Flor. ed. 3, appendix p. 535. A. Bennett, Ex. Club Report, 1895, p. 496 (1897). Fryer, l.c. p. 497. P. sparganifolius, “ Bab. Man. ed. 6, p. 363 (ew parte) non Liist! nec Fries!” Syme, EH. B. vol. ix. p. 31. Bab. Man. ed. 7, p. 372. Lond. Cat. ed. 8, No. 1484, ed. 9, No. 1585. P. polygonifolius, c, Kirkii, Lond. Cat. ed. 7, No. 1219. Sprcrric CHaracter.—Lower leaves submerged, ‘‘ sessile,” very long linear membranous ‘‘ with several rows of greatly elongate-cancellate areolations along the midrib” ; upper leaves like the lower but shorter, lanceolate or oblanceolate, or very long stalked and floating, coriaceous, oblong, or elliptical, or lanceolate. Flower- spike barren, short, on slender long equal peduncles, borne alike on branches with submerged or floating coriaceous leaves. Probably a hybrid; between P. natans ¢ and P. polygonifolius 3 ? Description: Rootstook not known. Stem terete, much branched, ‘‘ the lower branches barren.” Leaves : lowest submerged sessile (or stalked?) 6-15 in. long by 1} in. wide, membranous, pellucid, bright green, attenuated gradually into the base, more abruptly so at the acute apex, entire, with 8-14 lateral ribs with distant irregular transverse veins ; midrib bordered by several rows of elongate cancellate areolations forming a more or less conspicuous band along the middle of the leaf (the leaves decayed by the time of flowering ?). On the upper fertile branches the leaves are of two kinds, neither of which is essential to the production of flower-spikes, usually submerged and of the same structure and substance as the lower leaves, but more distinctly stalked, the lamina passing insensibly into the petiole, clustering, flabellate ; floating leaves coriaceous or subcoriaceous, very long stalked, oblong or elliptical, lamina rounded and abrupt at the base, sometimes slightly auricled, or lanceolate and gradually narrowed into the petiole, lateral ribs opaque, midrib slightly dis- coloured and prominent at the base as in P. polygontfolius, and all the leaves with areolations as in that species. PrttotE 38-5 inches long, greatly exceeding the lamina, somewhat herbaceous and jndistinguishable from the lamina in the lower, and lanceolate upper leaves. Strpuues “very long, blunt, not winged on the back,” but with two distinct green ribs forming a decided keel, ‘‘ scarious with numerous fibres,’ or herbaceous. SPIKE cylindrical, densely flowered, sepals not expanding, $ in. long, flowers barren. PrpuNCLE 3-4 in. long, slender, equal. Cotour light green, upper leaves darker green, drying darker, “upper leaves tinged with reddish brown.” P. KIRKII 17 Hasit: apparently with hnear lower leaves first springing in tufts like the phyllodes of P. nztans, sueceeded by greatly branched stems with or without coriaceous floating leaves on the upper branches; the flower-spikes are produced as in P. heterophyllus, independently of the coriaceous leaves. I have copied much of Dr. Syme’s original description, because it would be difficult to improve it in the absence of living specimens, which unfortunately neither of us could obtain. A root grown by anyone accustomed to study living plants of Potamogeton would solve much that is now obscure in the life history and relations of this remarkable form. Probably it is a hybrid; it seems to fall naturally under the fluitans group, from some of which some of its states are hardly distinguishable; the lower part of our plate 5, P. crassifolius forma verruta, might almost pass for the form of P. Kirkit figured in plate 8. I regret we could not figure a flowering example of this state, which is well drawn in Syme’s figure, EH. Bot. ed. 8, plate 1403. For a long time English botanists erroneously considered P. Kirkii to be the P. sparganifolius of Laestadius ; this mistake arose from attaching too great import- ance to the long narrow lower leaves; these, when dry, are often very narrow and perhaps more strongly suggestive of P. sparganifolius than those figured in our plate. When we compare the mature plants we find the flowers and spikes are very distinct, those of the latter plant being broad and less densely flowered, with expanded sepals, and fertile, the lower leaves are brighter green, less numerous, not spreading like a fan ; and the floating leaves are long-lanceolate and less coriaceous (see Tiselius, Pot. Suec. Exsic. No. 5). Dr. Tiselius’ P. dubius (lc. No. 19) is much more like our plant, but, as he suggests, is rather P. graminens X natans than P. natans X polygonifolius. Other specimens from Dr. Tiselius, however, of P. dubius approach our P. Kirkii much more closely, but all seem to lack the peculiar lower submerged leaves. Mr. J. Baagiée sent me a beautiful form collected by himself in Denmark, with the name of ‘‘ P. fluitans forma rivularis Lange,” which seems to connect our species with P. fluitans. But it is always difficult, and often unsafe to judge from dried specimens, especially when not prepared in a similar manner. In Mr. Baagoe’s speci- men the leaves are nearly all reduced to short submerged phyllodes along the whole length of the stem, even up to the base of the terminal flower-spike. This remark- able form may result from P. fluitans growing in a rapid stream, or not improbably it may have a somewhat different parentage from typical P. fluitans of Roth. Dr. Syme (lc. p. 32) says of P. Kirkii, “1 have some suspicion that it may be a very luxuriant abnormal form of P. polygonifolius.” In addition to the reasons he then gives for doubting this, we may add the production of the flower-spike in the absence of floating leaves. dD 18 POTAMOGETONS Looanities ; At present this hybrid-species is only known from Ireland, where it was “found by Mr. Thomas Kirk in the Ballinabrack River at Maam, co. Galway” (Syme), ‘ Loch Neagh by Dr. D. Moore,” Lough Corrib, Galway, T. Kirk, Sept. 7th, 1854. Description of the Plates. Prater 8, upper branch with submerged leaves, this state sometimes produces flower-spikes, Maam, co. Galway, Coll. W. R. Linton, August 15th, 1885; Herb. C. Bailey. Puiate 9, flower-spike on branch with floating leaves, Maam, co. Galway, August 13th, 1885; Herb. E. F. Linton. Maam, co. Galway, Coll. W. R. Linton, August 13th, 1885; Herb. C. Bailey. Lower leaves, Maam, co. Galway, Coll. W. A. Shoolbred, July 5th, 1895; Herb. C. Bailey. 5. POTAMOGETON POLYGONIFOLIUS, Pourret. Act. Toulouse, iii. p. 325 (1734). P. oblongus, Viviani, Anal. Bot. 2 (1802); Cham. et Schlecht. Linn. p. 214, No. 19. Fries Novit. ed. 2, p. 29. Fieber, Pot. Bohmens, p. 19. Koch, Fl. Germ. vol. u, p. 775. Crepin, Fl. Belge Fragm. Fasc. i. p. 25. P. polygonifolius, Reich. Germ. vol. vii. p. 24, Tab. 44. Bab. Man. ed. 2, p. 371. Hooker, Stud. Flor. ed. 3, p. 481. Syme, E. B. ed. 3, vol. ix. p. 27, pl. 1400. Bab. Man. ed. 7, p. 371. Hooker, Stud. Flor. ed. 3, p. 481. Druce, Fl. Oxford, p. 282. Tiselius, Pot. Scand. Exsic. fol. 9, 10, 11. Speciric CHaractrer.—Leaves all stalked, lower submerged, membranous, upper floating, coriaceous or subcoriaceous; no linear phyllodes. Drupelets very small, yellowish-green when fresh, reddish when dry. Petioles of the coriaceous leaves often persistent after the decay of the lamina, and then falsely resembling phyllodes, but may be known by the lacerated end lacking the acute apex of a perfect leaf. Description: Roorstock slender to stout, contorted, with many short internodes, with numerous fibrous rootlets, decaying in autumn except three or four terminal — internodes, the terminal one bearing a small reddish winter bud; nodes annular, prominent. Stem terete, ascending at irregular distances from the rootstock, sometimes from every node, or at intervals of 3—4 nodes, rooting at the lower joints and then throwing out stolons. In shallow-water forms the stem is only one or two inches long, or obsolete. Leaves all stalked ; lower submerged, pellucid, membranous, gradually narrowed into the petiole, often absent in shallow-water forms; lowest rarely narrowly linear- lanceolate, but then being a linear lamina, not a phyllode; upper very long stalked P. POLYGONIFOLIUS 19 in deep water, short or long stalked in shallow-water forms, coriaceous, lamina suborbicular to lanceolate, cordate, or abruptly or gradually narrowed into the petiole ; midrib and principal lateral ribs somewhat prominent on the underside; midrib conspicuous, prominent and somewhat discoloured at base of the lamina on the upperside, but never narrowed into a joint below the leaf as in P. natans. PETIOLE coriaceous, or herbaceous, slightly convex above, in deep-water forms usually greatly exceeding, but in shallow-water forms often shorter than the lamina. Stiputes long, subfibro-scarious, not keeled nor channelled on the back, subpersistent, Spike short cylindrical slender, densely flowered and fruited. Pepounc.e slender, equal, many times longer than the spike. Fruit small, beak short, facial, dorsal margin rounded when fresh, keeled when dry with faint lateral ridges. Contour “ dull olive-green, often tinged with reddish brown” (Syme). Hasit in deep-water forms like that of P. fluitans; in shallow-water forms with creeping rhizomes producing short stems at regular intervals, with the leaves all similar and coriaceous; very variable, many intermediate forms occurring between these two extremes, which insensibly pass into one another. This species was confused with P. natans by Linneus and most of the earlier authors, and by some of the later ones, as Bentham in his Handbook of the British Flora (1865). Perhaps this error arose in the first instance from the resemblance of the coriaceous floating leaves of some deep-water forms to those of small states of P. natans. The resemblance is merely superficial, and possibly the latter species approaches P. lucens almost as nearly as it does P. polygonifolius. When the Potamogetons of the whole world are better known I have little doubt that many species will be found falling in a natural arrangement between these two. The whole series of Australasian forms which have been assigned by authors to one or other of these two species will probably be found to belong to neither of them, but to con- stitute intermediate species. To evolutionists it will seem probable that the European flora has possessed and lost many such intermediates ; so really distinct and apart are the two species originally mixed in P. natans. P. polygonifulius is so variable according to the depth of the water in which it grows, and is so susceptible to the influences of local conditions of every kind, that each district, almost, seems to produce a peculiar form. Hence, as may be supposed, many striking or extreme forms have had varietal names attached to them; or, at least, some attempt has been made to arrange the forms in natural groups. ‘The best of these arrange- ments is that of the late Dr. Boswell Syme, in his excellent edition of Higlish Botany, and is as follows :— 20 POTAMOGETONS “Var. genuinus (Eng. Bot. pl. 1400). “ Lowest leaves narrower and thinner in texture than the upper ones, which are rounded or cordate at the base. “Var, pseudo-fluitans (Journal of Botany, Tab. 342). «“ Lower leayes membranous, elliptical-strapshaped, attenuated at each end; floating leaves subcoriaceous, gradually attenuated into the petiole. “Var. ericetorwm (our plate 11, lower figures). « Leaves all similar, subcoriaceous, floating or rising out of the water, longly- stalked, with an oblong-oval or roundish lamina.”’—KEnglish Botany, ed. 3, p. 28. Taken in their extreme forms the above “ varieties’? seem so natural and so distinct that we might suppose they could be advantageously retained in the present work. On looking at a good series of specimens, however, it will at once be evident that the intermediate forms are so numerous and so gradually link “ var. ericetorum”’ to “ var. pseudo-fluitans,” that more than half our series will remain unnamed. Under cultivation, by gradually increasing the depth of the water, “ var. ericetorum speedily becomes “ var. genuinus,” and the same change has been noticed in plants growing naturally. Mr. Druce, in his “ Flora of Oxfordshire,” writes (p. 283) :—* In 1885 I noticed the floating form in a small pond . .. and a month later, this being dried up, the plant assumed the heath form”; the reverse result here being obtained. The plant from the River Leven, Loch Lomond, selected by Dr. Boswell Syme as the type of ‘var. pseudo-fluitans,’ is probably the ordinary deep-water “ genuinus,”’ with leaves elongated and membranous (even in one upper leaf) by being kept almost continuously submerged by running water. I have not been able to obtain specimens to cultivate from this locality, but from observations made by Mr. Alexander Somerville, of Glasgow, on a very similar form found by him in the Isle of Arran, a form with “membranous lower leaves,’ and otherwise exactly agreeing with Dr. Boswell Syme’s description, has been traced through ,“ var. ee i) 9 genurmus almost or completely to ‘‘ var. ericetorum.” One more so-called ‘“‘ variety” has found a place in our British Lists, and so it must be briefly noticed here. It was named “var. angustifolius”” by Fries (Novit. Flor. Suec. ed. 2, p. 80). Ido not understand his desciiption, and therefore do not copy it. The specimens figured in our plate 11 were collected by Mr. Charles Bailey, and named by Mr. Arthur Bennett from type spc cimens of Fries’ plant. This form also passes insensibly into the other above-named ‘‘ varieties.” If we accept such transient states as “varieties,” we might as well name a dozen others equally indefinite. The form figured on plate 12 is not only distinct in appearance from the usual P. POLYGONIFOLIUS 21 states of polygonifolius, but it remains unaltered when cultivated in stagnant water, under conditions wholly differing from those of its natural station. It was found by Mr. W. H. Beeby in 1890; the following account of it is extracted from his paper on The Flora of Shetland (Scottish Naturalist, January, 1891). “Potamogeton (polygonifolius x heterophyllus?) Walls (district of). POTAMOGETONS ed. p. 54 (1828). P. filiformis, Nolte, Syme, Eng. Bot. ed. 3, ix. 55, t. 1424 (1869). Hooker, Stud. Flor. ed. 8, 436 (1884). Morong, Naid. N. America, 50, t. 57 (1893). Babington, Man. Brit. Bot. ed. 9, 448 (1904). Ascherson u. Graebner, Synop. Flor. Mitteleurop. ed. 2, 542 (1918). Description. Creeping in the mud with slender rhizomes, and throwing up tufts of leaves at intervals. Srems slender, terete or subcompressed, filiform, sparingly branched at the base, and simple above. LEAVES all similar, setaceous, subacute, 1-ribbed. STIPULES rather long and narrow. PEDUNCULES equal 4—9 inches long. SPIKES 2 inches, interrupted, few-flowered in whorls. SEPALS obovate. Fruit. Drupelets small, olive, subcompressed, semioblanceolate on the ventral margin, semicircular on the dorsal, bluntly 3-keeled, with a short, nearly central beak. Plant bright green, generally turning darker in drying. When not in fruit and robust marinus is difficult to separate from pectinatus, which at times assumes a jiliformis-like aspect, but when in fruit specimens can be easily separated. DiIsTRIBUTION in the British Islands. Anglesea (Davies herbarium), Berwick north to Shetland, Outer Hebrides ; in about 17 counties. Ireland: Galway to Armagh ; in 10 of Mr. Praeger’s divisions. DISTRIBUTION 1n other countries. Europe : Iceland, Faroes, Norway, 70° 51’ N. Lat., Sweden to Piedmont, France (H. Alps), Russia. Asia: Siberia, 70° N. Lat., Pamirs and Mongolia to Turkestan, Persia, India, Japan. Africa: Egypt, Reuwenzni Expedition, 1893-4. N. America: Greenland to 71° N. Lat., Canada, Labrador to Hudson’s Bay and B. Columbia, U. States, Maine to Dakota and California. S. America: Chili (Gay); Argentina (Fries). Australia: Victoria, 8. Australia, New Zealand. EXPLANATION OF PLATE 60. PoramocGEToN MARINUS. Drawn from Forfarshire specimens ; with longitudinal section of fruit and section of stem, mag. 42. POTAMOGETON PENNSYLVANICUS, Cham. et Schlecht., Linn. ii. 227 (1827). P. Cuaytoniu, Tuckerman, Am. Jour. Science, Ser. 1, vol. xiv. p. 38 (1848). P. Nuttalii, Chamisso et Schlechtendal, l.c., teste Morong, in Naid. N. America, tab. 29, 2(1893). Ar. Bennett, Naturalist, x. 10, 373 (1908). Ar. Bennett, Trans. Edinb. Bot. Soc, vol. xxii. p. 211 (1908). P. PENNSYLVANICUS 91 Description. ‘Stems slender, compressed, mostly simple, 0-3-2 m. long. Floating leaves opposite, elliptic to obovate, obtuse, short-petioled, 3-8 cm. long, 8°24 mm. wide, many-nerved; submerged leaves linear, 2-ranked, 5-17 em. long, 2-6 mm. wide, 5-nerved with reticulations on either side of midrib; stipules obtuse, hyaline, not keeled; peduncules 2-13 cm. long; spikes 1-2-5 cm. long; fruit round- obovoid, 2°5-4 mm. long, 2-3 mm. thick, 3-keeled, the sides flat and indistinctly impressed ; style short, apical ; embryo coiled one and one-third _ times.” Morong, ex Britton, Man. Fl. N. States and Canada, p. 43 (1901). DistripuTion. From N. Scotia to B. Columbia, and Maine to Georgia, and New Jersey to Washington Territory. Found by Miss A. E. Vigurs in the canal at Salterhebble Bridge, near Halifax, Yorkshire, in 1907; and, the next year, between there and Elland, three miles up the canal. Most likely | P. Pennsylvanicus. introduced with cotton, as it is one of Fig. 2—1, plant; 2, flower; 3, segment of perianth with stamen ; 4, segment of perianth; 5, fruit; 6, ovary. the common species in the United States, being abundant in the States where cotton is grown. I believe this to be the only instance of the introduction of a Potamogeton. The supposed case of P. crispus, L., to North America proved a mistake ; although said to be “ not known to the earlier American botanists,” Pursh recorded it in 1814 for several habitats, and says he has seen living specimens, “ Canada to Virginia.” I give the earliest name of which there are specimens extant (i.e. at Berlin), but it is probable that Rafinesques’ name of P. epihydrum in Medical Reports 2nd Hex. vol. v. p. 354 (1808) will eventually have to be accepted. Dr. Morong, l.c., adopts the name P. Nuttalii or the page before the one I quote, . but while well knowing the keenness and ability of my late friend, I cannot agree with him in this, though of course I may be mistaken. CORRIGENDA. Page 47, headline, for “ Cooperi” read “ crispus.” Page 56, line 1, “densus” should have been numbered 16, instead of 17, the number of subjects separately described in the work is 41, and not 42 as numbered. . The subjects on Plates 39 to 47 do not follow the same order as in the text. But the references to the Plates in the text are correct. Plates 20 and 26. Figures 6, 7, 8 on Plate 26 are the fruit of P. alpinus omitted on Plate 20. ea -r X . Names of groups are printed in small capitals, and of synonyms in italics. The page on which the main description of the plant is given is indicated by black numerals. In this Index a x is used to indicate the species crossed and not the name of the hybrid. acuminatus, v. (pusillus), 84. | acutifolius, 79. Plate 51, alpinus, 27. ——— Plates 18-20, 26. x lucens, 76. x natans, 31. xX polygonifolius, 31. xX praelongus, 34. angustifolius, 72. angustifolius, v. (polygonifolius), 20, 22. australis, v. (pusillus), 82. - Bennettii, 47, 53. Plate 33. Bertcholdii, v. (pusillus), 84. Billupsu, 25. Plates 16, 17. borealis, f. (alpinus), 28. | caepitosus, 82. caneellata, f. (polygonifolius), 21. Chloephylli group, 2, 57, 77. Claytonii, 90. CoLEOPHYLLI group, 2, 57, 86. coloratus, 21, 22. Plates 13-15. x Zizi, 25. compressus, 77, 80, 81. Cooperi, 47, 48. Plates 31, 32. coriaceus, 27, 59, 74. Plate 43. | x natans, 8. x perfoliatus, 69. cornutus, 72. v. (crispus), 45. crassifolius, 8, 15. Plates 4, 5. f. (verruta), 9, 17. | falecatus, 68. | filiformis, 90. | fluitans, 11. | fluitans, 28, 31. crispus, 43. Plates 29, 30. x Friesii, 54. x obtusifolius, 53. xX perfoliatus, 48. xX praelongus, 49. curvifolius, 59, 69, 70. | cuspidatus, 79, 80. cymatodes, 48. decipiens, 59, 75. Plate 48. v. salicifolius, 76. densifolia; f. (perfoliatus), 39. densus, 56. Plate 34. dichotamus, v. {pectinatus), 89. Drucei, 31. Plate 21. dubius, 17. elegans, f. (praelongus), 36. ENANTIOPHYLLI group, 2, 57. epthydrum, 91. ericetorum, v. (polygonifolius), 20, 22. eupectinatus, sub-sp. (pectinatus), 86, ; Plate 40. flabellatus, sub-sp. (interruptus), | 88. Plates 6, 7. f. rivularis, 17. fluvialis, v. (obtusifolius), 81. jluviatilus, v. (pectinatus), 87. Friesii, 81. Friesii, Plate 53. x crispus, 47, 54. x heterophyllus, 63. gemmifer, v. (crispus), 43, 47. genuinus, v. (alpinus), 29. v. (polygonifolius), 20. gracilior, f. (alpinus), 28. gramineum, 59. gramineus, 64, 65, 80. x nitens, 68. graminifolius, 64. Plate 36. GRASS-LEAVED section, 77. Griffithii, 34. Plates 22, 23. HETEROPHYLLI group, 2. heterophyllus, 59, 65. Plate 35. x Friesii, 63. x polygonifolius, 21. x pusillus, 63. 64, ——— f, (lucens), 72. sub-sp. (nitens), 70. v. (perfoliatus), 48. hibernicus, v. (lanceolatus), 63. HomopHYLLI group, 2. TT | homophyllus, v. (alpinus), 29. Hornemanni, 22. imbecilla, f. (perfoliatus), 39. interruptus, 88. Plate 59. involuta, £. (nitens), 69. involutus, 69. Plate 44. | iterata, f. (alpinus), 28. Jacksoni, v. (perfoliatus), 48 junctifolius, 87, P 4 Kirkii, 10, 16. Plates 8, 9. lanceolatus, 63. Plate 39. linearis, v. (natans), 6, lonchites, 13. longifolia, f. (alpinus), 28. Lucens group, 59. lucens, 71. Plates 45-47. f. alpinus, 30. . 7 Vv. coriaceus, ai. heterophyllus, 72. sub-sp. Zizti, 72. x alpinus, 76. x natans, 11, 14. x perfoliatus, 75. major, 81. marinus, 89. Plate 60. marinus, 89. monogynus, 85. mucronatus, 81. natans, 4, 19, 24. Plates 1-3. x alpinus, 31. x coriaceus, 8. x lucens, 11, 14. x polygonifolius, x Zizii, 8. natans, Vv. prolixus, 6. nitens, 7, 59, 69. ——_— Plates 37, 38. f. involuta, 69. X gramineus, 68, Nuttali, 90. oblongus, 18. obtusifolius, 80. Plate 52. -——— x crispus, 47, 53. - y. (perfoliatus), 41. sub-sp. decipiens, 75. INDEX, pachystachyus, v. (plantagineus), 22. : palustris, f. (alpinus), 28, pectinatus, 86. Plate 57. sub-sp. flabellatus, 88. v. fluviatilus, 87. Pennsylvanicus, 90. perfoliatus, 38, 59, 70. Plates 27, 28. v. heterophyllus, 48. v. Jacksoni, 48. x coriaceus, 69. x erispus, 47, 48. x lucens, 75. Phialz, 85, plantagineus, 21, 22. polygonifolius, 17, 18, 24. Plates 10-12. f. cancellata, 21. v. pseudo-fluitans, 33. x alpinus, 31. x natans, 16. polygonifolius, 16. ce. linearis, 6. praelongus, 36. Plates 24-26. x alpinus, 34. xX erispus, 47, 49. prolivus, 6, Proteus, 59. curvifolius, 69. heterophyllus, 65. -——— lucens, 71. Zizi, 72. pseudo-fluitans, v. (poly gonifolius), 20, 33. pusillus, 83, 84. Plate 55. v. australis, 82. x heterophyllus, 63. purpurascens, 28. 16. rigidus, v. (pusillus), 84. Robbinsii, 55. rotundifolia f. (perfoliatus), 39. rufescens, 28, rutilus, 82. Plate 54. salicifolius, 76. salignus, 76. Plate 49. scoparius, v. (interruptus), 89. serratum, 56. serratus, 45. serrulatus, 55, v. (crispus), 43. similis, v. (pusillus), 84. sparganifolius, 16. spathuleeformis, 67. spathulatus, 33. stagnatilis (fluitans), 11. Sturrockii, 84. tenuissimus, v. (pusillus), 84. trichoides, 85. Plate 56. Trimmeri, v. (trichoides), 86. tuberculatus, 85. undulatus, 47. undulatus, 48. v. Cooperi, 48. vaginatus, 87. Plate 58. Vaillantii, 89. varians, 67. Plate 41. verruta, f. (crassifolius), 9, 17. virides, f. (alpinus), 28. vulgaris, v. (pusillus), 84. Zizii (= Angustifolius), 59, 72. Plate 42. Zizi II. coriaceus, 74. Zizi < coloratus, 25. — xX natans, 8. zosteraefolius, 77, 80. Plate 50. LONDON: PRINTED BY WILLIAM CLOWES AND SONS, LIMITED, DUKE STREET, STAMFORD STREET, S.E., AND GREAT WINDMILL SIREET, W. PLATE 3. os Lot POLL LE in ott al i ee Vincent. Brooks, Day & Son imp R-Morgan del et lith L Reeve & C° London -° London L Reeve & PLATE 5 al Vincent. Brooks Day& Sonknp. R Morgan delet hth= — L Reeve & C? London RMorgan del eLlith. & C2 London Reeve L PLATE 7. RMorgan delet lth. Vincent Brooks Day & Son Imp L.Reeve C° London R Morgan delet lith. Vincent Brooks,Day &Son Tm, L Reeve &C® London. Vincent Brovks,Day & San imp. R Morgan del ¢& ith L Reeve &C° London RMor$an del et lith Vincent Brooks Day &Sonk § f: to) 4 ° 12) oe & 4 PLATE Il SIVA Vincent Brooks Day & Son imp R Morgan. del, et hth LReeve & C°London. PLATE 12 Vincent Brooks,Day & Son Imp — WY ie ~ We ib Se "emis fi) fag et Spek! fer! = ee SS ae ee, TO CL eae RS SF Sr L Reeve &C° London SSS « ‘ ~ x " ~s = . +e 22 sabe? > ——— .2] R Morg opps ; gan del.et lith PLATE 14. PU wae: Jie ry Cee ys ding hs “i eee a Wrage oH it Lf PRs LOD, vn { OR y | ‘wit : | Ny th, SEN " et Bn Bae (as UU f ; 7 i ae ig i Min ri ithe . ag re a - wis sh aa Mm Lyell eal KS Aiwa, mas cee SRNR ; a NUNN ARAN ee *~ Se we RMorgan del et hth. : Vincent Brooks, Day & Son Imp. LReeve & C° London. 3 ; e : 8 3 : e “ r= 3 g a § & g d 8 Pa x oO 3 Ey @ o | BR. Morgan deeb Hui. Vincent Brooks, Day & Son Imp L Reeve C° Landon R Morgan del,et-lith Vincent. Brooks Day & Sonim RMorgan del, et hth. L Reeve C° London PLATE19 Vincent Brooks ,Day & Son Lt*Imp RMorgar delet hth ul L Reeve &C® Londort 4 af 7 ty fy BR A ks Day & Son Lt“Imp sn VDOroo, Vir f L Reeve &C°Landon - c ~AS WW ww ‘ ike ~~ aA Ls, a AN SAY SAAS s, N\ SQA mA SAAR & =o yA th vase R Morgan delet lith E Londaz PLATE 22. Ay fj ® - a a / VancentBrooks Day &Son Ii4 Imp. L Reeve &C° London. L Reeve & C? London. PLATE 23 — ee ed — Vincent. Brooks Day & Son Lt# imp “RMorgoan del, et Jith. n Lt? Img Brooks Day&S Vincent na ote L Reeve & C° Lor Vincent Brooks, Day &Son Lt*h L Reeve & CS London “al 97 PLATE g : $ : Beet Vincent Brooks,Day &Son Lt Im don L Reeve & C° Lone PLATE 28. Vincent Brooks,Day & Son Litimp. RMorgan del et lith L-Reeve & C° London L Reeve & no — London Vincent Brooks Day & Sa 90) PLATE -RMorgan del et ith Vincent Brooks Day &SonLtimp L Reeve & C2 London. 3l PLATE It Imp Son Day & Vincent Brooks R-Morgan del, et hth i.don (8) NOT teeve & CO] TPR 1. tN ai i SPT Rts Sky. ee eA Hee ears bE T \ ‘ > Be sz he! [7 ( ih) Oe An ae eee Se eS ee LO ALIGN a RMorgan del et hth. dae o~ Oe C° London LReeve & PLATE 33. OT it IO a wee . RMorgan del, et lith. Vincent Brooks Day& Son Lt Imp LReeve &C° London. 3 = fe Oo, Vincent Brooks Day & SonIt4inp, L Reeve & C®° Landon. PLATE 3 : j Vincent Brooks Day & So RMorgan delet ith. 7 | ' L.Reeve &C° London PLATE? Gi : Vincent Brooks Day & Son Lt# | ee , et “Vas ewe : re ea L Reeve &C° London. a Vincent Brooks, Day & Son Lt?1 R Morgan del, et lith L Reeve & C° London PLATE 38 a RMorgan del et hth Vincent Brooks Day & Son Lt* Imp L. Reeve & C° London PLATE 39 Vincent Brooks Day & Son Ind Imp. L. Reeve & C® London. PL OS — ee Vincent Brooks Day &Son Lt* imp R Morgan. del,eblith aan Lon LReeve &C® PLATE 41 i4 ; Vincent Brooks. Day & Son Lt* imp y . 7 x : | ‘| Pa ; ; 7 ae ated ( f 2 Ryn Beet p- ayo vag eee aa asd = b., ) SA, ; jy, ; Art cal vw L Reeve & C?2 London Voy. dpe PLATE R Morgan del. etlith Vincent Brooks Day & Son Lt* L. Reeve & C° Landon VEEP ma « acien — ——— Lat hn oT Vincent Brooks Day &S ir, an del. eb th 1 pV ite) R — ondon &CeLl Reeve du PLATE 44 Vincent Broaks Day & San Limp RMorgan delet ith PLAT Vir Le - . mn ey te uke D2 Og OO @: tas@™ oF MAS eae ahtes oe eo Pla 5 ey at ~ S < > ~ R Morgan del et hth Vincent Brooks,Day &San Lt* inp R.Morgan del et. th. c ° vU S ts) 4 ° 1@) % © > v ® X 4 r th hh + et del san he : Mi ma PLATE 48 @ IO eee ox 98, 3 , R Mongan del,ethth : Vincent Brooks Day & Son Lt imp L.Reeve & C° London cent Brooks, Day& SonLt*imp Vin E CKmght del.et hth C? London PLATE 50. E.C Kmght delet ith. . be Day & Son Litimp = L-Reeve &C° London PLATE 51 E.C.Knight del et hth Vincent Brooks, Day &Son Lt? mp L Reeve & C° London. PLATE 52 M.Snuth del et lth. c Vincent Brooks, Day & Son Lif imp L.Reeve & C° London. E.CKni ght delet hth PLATE 54. Vincent Brooks Day & SonLt?#imp M.Smuth del et Inth. L.Reeve &C° Landon. = PLATE 55 Vincent Brooks Day &Son Lt imp M.Smith de] et hth L. Reeve &C° London | \ RY | / \\) \ \ MY \ ee — | VV r VY H} 4 < — A | : j | I | ' ) | : " | y {2 z ; / | } \ f | Ey, Z | ; : * y Wy Vf f___—=ff f : yf A, Vp 7 é | } Sy, / | ee 424i, y, : \ \ ~ / \\ / SF =. as a} nN y, / ! Yr g j “A PLATE | son Lt@ amp. es) wu a or £ c ba eal taal 2 =] - 6 ‘3 + O x z fa R Margan del 0 2400 80 Zt v0 2 Wall JON I lV WL