Biomed UK 520 F398 1900 8 I 7 4 UNIVERSITY OF CALIFORNIA AT LOS ANGELES FERNWORT PAPERS, PRESENTED AT A MEETING OF FERN STUDENTS, HELD IN NEW YORK CITY JUNE 27, 1900, UNDER THE AUSPICES OF THE LINN^AN FERN CHAPTER. Issued December 20, 1900. Printed for the -Linnaean Fern Chapter. BlNGHAMTON, N. Y. WIZARD N. CI" slender, the submersed ones i5-4ocm. long, ascending, in spirals, stomata few : spores 400-480 ft with slender spinules often mixed with short or elongated crests. After the fall of the water in summer, the long leaves disap- pear, and are succeeded by short, bright green ones 5-7011. long. In this state the plants would not be taken for the same unless the transition had been observed. There is a specimen from Boott's herbarium, collected in this stage, in the National Herbarium. Maine : Moosehead Lake, Harvey; various localities, Fernald. New Hampshire : Kingston, East Kingston, Newton, Epping, New- market, A. A. Eaton. Vermont: Norwich (fide Dodge). Massa- chusetts : Woburn creek and Abajona river, Boott. Connecticut : Groton, Graves. 9. I. EATONI Dodge. The largest New England species, exceeded in size by none now known in the world, and equalled only by Engelmanni valida and the European Maliaverniana. Leaves 25-200, as much as 6ocm. in length in water, about io-i5cm. when growing on banks. Known at once in the field by its size, but especially by the sporangium, which is light brown in color ( white in nearly all other New Eng- land species), and is sparsely rilled with very small spores. Mic- rosporangia are rarely found in this species, but as it is abundant in several localities, it certainly must bear microspores, as it never multiplies by offshoots. It may be possible, but not probable, that some sporanges bear both kinds of spores. I have noted such sporanges in Tuckermani and several other species, but the micro- spores are usually aborted in such cases. Found thus far only in the waters of three small rivers Powow at East Kingston and Kingston, N. H., and Amesbury, Mass., in the Lamprey at Epping and Newmarket, A. A. Eaton; in Parker river at Georgetown, Mass., Dodge. 10. I. Gravesii n. sp. Plant dioecious or polygamous, rather large : rootstock bi- lobed : leaves 50-75, 12-15011. long, imm. in diameter in the mid- dle, erect, sharp pointed, dark green, with abundant stomata and four bast-bundles : velum quite narrow, inner sporanges oval, light cinnamon in color from the abundance of vermiform, translucent, light-colored sclerenchym cells ; macrospores many, small, 351-405 n in diameter, tetrahedro-globose, the upper facies flat, densely cov- ered with short, truncate, mostly single columns ; microspores not seen. Goshen, Conn., Underwood, 1899; gravelley tidal shore, Sel- dens Cove, Lyme, Conn., August 31, 1900, Dr. C. B. Graves. Specimens were sent to Dodge from this locality by Graves in 1895. It was at first referred to Eatoni, but was finally separated, and has since lain without name. It has the aspect of sac charata; \s\\\. its affinities are with Eatoni, with which it agrees in being polygamous, in appearance of sporangium and shape of spores, which in both have the appearance of being abortive. It is a smaller plant with erect leaves, while the emersed ones of Eatoni are not. The spore sculpture, resembling echinospora rather than Eatoni, safely sepa- rates them. Though the majority of the columns are single, each spore usually has a few connected into vermiform or horseshoe- shaped figures, but they are walls rather than wrinkles, and in this respect resemble Eatoni but remotely. The type is in my private herbarium ; co-types are deposited in the U. S. National Herbarium, in the herbaria of the Missouri Botanical Garden and the University of Minnesota. ii. I. DODGEI A. A. Eaton. Plants medium to large : leaves 10-75, the submersed spirally ascending, 2o-45cm. long, the emersed io-2ocm. long, often, espe- cially when the plants are not crowded, tortuous and interlaced, 2-3mm. broad : velum one-fifth to one-fourth indusiate ; sporangium spotted ; macrospores 500-675 /*, averaging 560 p, with mostly scat- tered groups of irregular, spinulose-rosulate crests ; microspores ashy, 22-40 /u, wrinkled. Abundant on the banks of Powow river where overflowed for the purpose of forming a pond during the greater part of the winter, at Kingston, N. H., the only known New England locality. 12. I. ENGELMANNI A. Br. Leaves 15-100, bright green, usually erect except when grow- ing out of water in bare places, stomata abundant : sporangium un- spotted, velum narrow ; macrospores chalk-white, 350-550 /i ftli.r-foemina. In the course of this study it seemed natural that I should take up also the question of the genus of these ferns ; and having done so, I propose to. lay the results of that study before the Fern Chapter, as briefly as the subject will permit. Although all writers on ferns are obliged to recognize Athyrium by the character of its sori, the tendency has been, since the publi- cation of Hooker's Species Filicum, to make it a section or sub- genus of Aspleniuin. Previous to Hooker's time. Roth's genus Athyrium had been quite generally accepted by German and even by English botanists ; and since Hooker's time, Thomas Moore, the most careful monographer of the English ferns, has retained Athyrium for filix-foemi na and its varieties. In our own country, since the time of Pursh, who followed the Linnaean nomenclature, and placed these ferns under Polypodiutn, and Dr. Jacob Bigelow, who adopted the nomenclature of Swartz and regarded them as 26 forms of Aspidium, American botanists have accepted Hooker's \-ie\vs. Because the sori are long and the indusia are attached by one side to the veins and open on the other side, all of which fea- tures are characteristic of Asplenium, it has been claimed thaty?/T.r- foemina and other species allied with it belong to that genus. But there is another character which essentially modifies this judgment and compels recognition. That is the curvature of the sorus and indusium, by which the upper end of the sorus is curved across the veinlet and back upon itself, often so strongly as to resemble the sorus and indusium of a typical Lastrea, or free-veined form of \ephrodium, distinguished by its kidney-shaped indusium. It was this roundish form which induced Linnaeus to rank it as a I'olypo- dium, and Swartz to place it under Aspidium. These three genera therefore have laid claim to it, with each of which it agrees in a modified degree, but with no one of which it agrees entirely. The genus Athyrium, however, as constituted by Roth, comprises all these features, and completely satisfies all the generic conditions dependent upon its fructification. What those conditions are will best be discovered by taking Roth's own description of the genus, which I have translated from the Latin : ATHYRIUM. " Capsules distributed in ovate sori underneath the disc of the frond, surrounded with an articulate ring. The involucre springs laterally from the venule, lying loosely in the form of a scale, with laciniate-fimbriate margin, at length elevated inwardly, pressed back and semi-lunar. " Observation. The essential character of Polystichum consists in the involucre being either umbilicate or peltate, or reniform, but on every side nearly free. In the first case, at the time of maturity, the involucre is drawn together centrally to its own fixed point, and very often acquires the shape of a funnel ; but in the latter case it is drawn back sublaterally to a fixed point and changes the sub-peltate shape into a reniform shape. " But among the Linnaean Polypods there is observed no other of which the involucre has an ovate-oblong shape, and as in Asplen- iniu, springing laterally from the venule, draws itself out following the length of the same. It lies more loosely in a heap before the maturity of the capsule, when from the other side opposite and looking backward it rises a little above the costa of the frond or of the lacinia. Toward maturity, it is raised against the costa by the inward increase of the capsules bursting forth, by which means it assumes a semi-lunar shape. Influenced by these reasons, accord- ing to the method of founding the genera of ferns upon the invo- lucre, I have come to the conclusion that plants of this nature should be removed from the rest of the Linnaean Polypods, on account of the situation, shape and condition of the involucre. Nor can they be included in Asplenium, on account of the plainly distinct condi- tion of the involucre, although they approach nearer to this genus. Therefore it seems necessary that I should establish a genus of their own, to which I have given the name Athyrium.'" Now it will be seen from this description that Roth distinctly differentiates Athyrium from Asplenium. He places only seven species under the genus, of which A, fontanum and A. Halleri, now recognized as forms of one and the same species, are placed first. In this he follows the universal custom of placing at the head of the list the smallest and simplest forms, not because they are the most typical species of the genus, but because they are the smallest. The next five species are all now recognized as varieties of A. filix- foemina, and these are the ones which conform most clearly to his description of the genus. In commenting upon this genus, I would say that more or less of the sori upon each frond are generally straight ; some are ham- ate, like a shepherd's staff ; while in other varieties they are bent double and become hippocrepiform or almost round by age. If we ignore the genus Athyrium, it is easy to see how one authority can be justified for \AajCttvgfiKx-foetnina in Asplenium, while another is equally justifiable in calling it an Aspidium. It has seemed to me that the wiser course is to accept Athyrium as a valid genus, fol- lowing the example of Roth and Newman and Moore, the men who have made the closest study of the genus and the most detailed ex- amination of this particular species. Another character of the genus which is distinctly laid down in Roth's description of it, is the " laciniate-fimbriate margin" of the involucre. There is a difference of opinion among authors as to the exact value of such a character as this, even when it is constant. That it is not constant, can be certffied to by all who have examined any considerable number of specimens of our American A. filix- foemina. The free edge of the indusium is generally irregular and somewhat lacerated, but it could not properly be called " laciniate- fimbriate." But the case is different with Athyrium cyclosorum. Perhaps no better evidence of the validity of the genus and of 28 cvclosorum as a species could be found than by taking a vomit; frond of A. filix-foemina rubellum just after the laciniae have un- folded in the spring, and a frond of A. . cyclosorum at the same stage of growth. In rubellum the sori are seen as distinctly as when mature, and even at this early period they are beautifully hamate, and the outer edge of the indusium is simply a little rag- ged. That is all, and there is hardly enough of the raggedness even to be noticeable. On the contrary, the sori on the just open- ing fronds of A. cyclosorum strictum are more often hippocrepi- form than hamate, and are sometimes nearly umbilicate, as if they were drawn together with a puckering string on the side attached to the vein ; while the outer edge is thickly fringed with long cilia plainly discernible to the naked eye, and which under a magnifier are seen to be many-jointed. While these facts show the nature of the genus in both cases, they also show that the fringed indusium can hardly be regarded as a generic distinction, but that in the case of A. cyclosorum it forms a very beautiful instance of specific dif- ference. If then we allow that Athyrium is a valid genus, the question naturally arises, " What species shall be admitted as belonging to the genus?" I should answer : all species which conform to Roth's description of the genus ; that is, all species endowed with a greater or less number of hamate or hippocrepiform sori and indusia, that being the character upon which Roth relied in distinguishing the genus from Polystichum and Asplenium.. Sir William Hooker, in his Species Filicum, threw discredit upon Athyrium by including with it as a sub-section, Robert Brown's genus Allantodia, which was founded on species having a swollen indusium and a straight, sausage-shaped sorus, as the name indicates. He also stated that "those who maintain the genus (Athyrium} are by no means agreed as to the species that should be included in it." That was not a fair statement. If he were going to judge the genus fairly he should have taken it as its author formed it and not as others had made it by introducing char- acters which the author did not use and evidently did not intend should be used in the genus. Hooker himself was one of the worst of these, for in his characterization of Athyrium as a sub-genus of Asplenium, he says: "Sori generally short; involucres lax, con- vex, straight, or often more or less arcuate and even hippocrepi- form, sometimes with the lobes unequal." He then includes A II- antodia as a section of Athyrium, giving as its character, " Invo- 2 9 lucres quite terete, very membranaceous, tender and brittle, often bursting irregularly." So far as I can judge, this is purely gratui- tous on Hooker's part, and was entirely foreign to the original idea of Roth. In conformity with that idea, no species possessing habitually straight sori has any place in the genus. Every one of the species and varieties which he names has more or less arcuate sori ; and Hooker's unfairness was accentuated when he placed A. fontamim under Eu-Asplenium, but described the involucres as "very small, athytioid" thus showing that he recognized the athyrioid involucre as a valid type. In conclusion, there are two or three points which I would like to bring out and make prominent. First is the fact that every one of the species which Roth originally put into his genus Athyrium was taken from the Linnaean genus Polypodium, and was never included by Linnaeus in the genus Asplenium, in fact was not con- sidered as having any relation to Asplenium. Secondly, Presl in his remarks on the tribe Aspidiae and the manner in which it was broken up by different authors, said : " First came Roth, who in the third volume (1800) of the Flora Germanica divided the indusiate Polypodia of Linnaeus into more genera, viz. into Athyrium, Polystichum, and Cyathea. The dis- tinguished Bernhardi drove Athyrium into Asplenium (the word Presl uses is repulsit, which is much stronger than removit would have been), accepted Polystichum, but changed Cyathea into Cys- topteris, on account of another genus so called by Smith." So we see that until Bernhardi "drove" Athyrium into Asplenium, ' the species which composed Athyrium had no relation with Asplenium whatever. Thirdly, Roth himself, as I have already shown, distinctly dif- ferentiated Athyrium from both Polystichum, which was his own genus, and the Asplenium of Linnaeus, and asserted that the char- acter of its sorus and indusium entitled it to be separated from these genera and to have a genus of its own. These points seem to me not only to establish Athyrium as a \-alid genus, but to separate it wholly from any of the other genera with which it has hitherto been associated. 30 ON THE OCCURRENCE OF THE HART'S=TONQUE IN AHERICA.* Bv WILLIAM R. MAXON. The following account comprises chiefly an historical narrative of the discovery of the American stations for the Hart's-tongue, t I'hyllitis Scolopendrium (L. ) Newm., \ with descriptions of habitat, especial mention being made of the natural conditions which appear to determine a suitable environment for the fern. Attention has been given also to the question of the fern's former distribution, and an attempt made to determine the principal causes that have operated to effect the peculiarly limited distribution of the present. I shall discuss the American stations substantially in the order of their discovery. Considerable care has been exercised in sup- plying in full and verifying all available references bearing directly upon the subject. The central New York stations are the only ones w r ith which I am personally familiar. These I have frequently visited at various times from 1895 to 1898, while living at Oneida and Syracuse, N. Y. To the several correspondents, mentioned later, who have generously furnished data, largely the result of personal observation, I would extend my sincere thanks. || CENTRAL NEW YORK STATIONS. In central New York the Hart's-tongue has been found grow- ing in four separate localities : (a) Geddes ; (b) Chittenango Falls ; (c) Jamesville ; (d) Perryville. The range of territory covered is comparatively small, and the stations are separated by short dis- tances only. Each station is entirely distinct, however, and scat- tering plants seem not to occur in the intervening- territority. * Published by permission of the Secretary of the Smithsonian Institution. tThe term is applied here only to the species Scolopendrium. \ Principal synonomy : Asplenium Scolopendrium L. Sp. PI. 1078. 1753. Scolopendrium vulgare J. E. Smith, in Mem. Acad. Tur 5: 421. 1793. Scolopendrium officinarum Sw. in Schrad. Journ. Hot. 2- Part 2, 61. (1800) 1801. Phyllitis Scolopendrium (L.) Newm. Hist. Brit. Ferns, Ed 2. 271. 1854. Scolopendrium Scolopendrium (i,.) Karst. Deutsch. Fl. Ed. i. 278. 1880-83. II A large part of the present paper is extracted from aa unpublished thesis by the writer presented to the faculty of Syracuse University, June, 1898, en- titled : " A Contribution to the Biology of the Hart's-tongue Fern." a. The Geddes Station. The Hart's-tongne was first discov- ered in America by Frederick Pursh, on July 20, 1807, near the place now known as Split Rock, a small suburb of Syracuse, about five miles west. Apparently the first mention in literature of this, the earliest American station, is contained in Pursh's Mora,* in which the following statement occurs: "In shady woods, among loose rocks in the western part of New York, near Onondago, on the plantation of J. Geddis, Esq. Perennial. July. v. v. This species I have seen in no other place but that here mentioned, nei- ther have I any information of its having been found in any other part of North America." Subsequent search failed to again dis- cover the fern in this locality, though it was found at Chittenango Falls (about 1830) and in the Jamesville vicinity (1857). The latter discovery doubtless led to the renewed attempts to find the plants at Geddes. At the suggestion of Dr. Asa Gray, Mr. J. A. Paine, in June, 1866, visited the locality for the purpose of verifying Pursh's origi- nal station. In his account of the trip f he says, after describing a fruitless search : " Hon. George Geddes, son of the J. Geddes, Ksq., referred to by Pursh, was then appealed to for information in general respecting this fern or its earliest station, and he readily cleared up the whole mystery. The place where it was discovered, he said, was nearly five miles west of Syracuse and half a mile south of his father's house ; on the single point of its being on his father's farm Pursh must have erred ; but it was nearby along a high ledge and about a celebrated sulphur spring." Paine did not succeed in rediscovering the fern at Geddes, and it was generally supposed that it no longer persisted there, until in 1879 (September 3oth ) it was rediscovered in fair quantity upon the Geddes property by members of the Syracuse Botany Club. ; The ferns continued to grow thriftily until the summer of 1895, when the Solvay Soda Ash concern blasted out the rocks which had so long served as a shelter. It is extremely unlikely that a single plant has survived. It seems strange that up to 1879 no one should have rediscov- ered the fern here. It may be of interest to note that the supposi- tion of Mr. George Geddes and Paine that Pursh was in error in his statement of its occurrence upon the farm is not borne out by * Pursh, Fl. Am. Sept. 2 : 667. 1814. t Am. Journ. Sci. & Arts, II. 42: 282. 1866. J Hull. Torr. Bot. Club. Q: 345-7. 1879. 32 the facts of its discovery in 1879. Mrs. L. L. Goodrich, of Syra- cuse, a relative of the Geddes family, and a member of the discov- ering party, has stated positively to me that the fern was upon that occasion found in considerable abundance well within the limits of the original "Geddes plantations." Pursh further says of this station : * " Mr Geddes brought me to a deep valley about i. m. from his house, where we ascended a steep very rocky hill ; here large masses of rock seem to be piled up or turned over one & another in such a confused manner, that it has left large chasms between them, which sometimes appear like caves : as it has a north aspect & overshadet with trees, all the rocks are covered with moss and vegetables : * * * & what I thought most of Asplenium Scolopendriuni this fern which I dont find mentioned by any one to grow in America I allways had a notion to be here ; & indeed I was quit enjoyed to find my preju- dice so well founded in truth. It appears Jo be the same as the european only smaler ; query ? is the european auriculated at the base like this species?" Split Rock is a limestone formation, consisting mostly of the Lower Helderberg, which extends to a depth of approximately 125 feet, with the so-called Oriskany sandstone interposing in a very thin sheet, from one to five inches thick, between it and the Cor- niferous of the Upper Helderberg (12 to 14 feet thick) which caps the plateau. Dr. Underwood has several times t called attention to the fact that in central New York at least, this fern is found only at the outcrop of the Corniferous. It was, before the inroad of the quarryman, a steep ledge about, 150 feet high and half a mile long, somewhat semicircular in general outline. At the base of the cliff was a brook, with a sulphur spring upon its bank. It was at some little distance from this spring, I am told, that the plants grew, somewhat sheltered by rocks and trees, but to a large degree ex- posed to the sweep of the cold northwest winds. b. The Chittenango Falls Station. The second place in Amer- ica at which this fern is known positively to have been found, is Chittenango Falls, in -Madison county, where Mr. William Cooper found it about 1830. From 1830 to 1857 it was the only American * Frederick Pursh : A Journal of a Botanical Excursion in the Northern Parts of the States of Pennsylvania and New York during the year 1807. 63-64. Edited by T. P. James, aud issued 1869. pp. 87. t Underwood : Our Native Ferns, Ed 5, p. 6. 1896. Also in Britton & Brown, Illus. Flora North. U. S. and Can. \: 21. 1896. station definitely known, * and perhaps on this account Pursh's dis- covery has often been incorrectly assigned to Chittenango Falls. The Chittenango creek, flowing northward toward Oneida lake, here takes a double plunge of over one hundred feet, and has worn through the limestone a rough gorge of that depth and more. The sides are extremely steep, but debris and soils have so accumulated at the base that the unbroken ledges loom up only along the top of the gorge. It is just out from under these overhanging cliffs and among the broken fallen fragments of limestone ( mostly Cornifer- ous) on the left bank, that the Hart's-tongue grows, perhaps thirty-five or forty feet above the level of the stream, and three or four rods distant. The soil is moderately moist, but light, yielding, and very rich in leaf mould. Here, scattered along the steep bank for a distance of nearly a quarter of a mile from the falls, the ferns grow in the fairly dense shade of second-growth maples, beeches, birches and elms. Among its companion plants are Pellaea Stel- /eri, Asplenintn Kitta-nmraria, Cystopteris bulbifera, and Dryop- teris Goldieana. c. The Jamesville Locality. The Hart's-tongue grows abund- antly in a number of places in the immediate vicinity of Jamesville, the distance included being ten to thirteen miles west of Chittenango Falls, and four to nine miles southeast of Syracuse. Exceptionally- fine plants grow in this locality, and almost the same conditions obtain in each of the stations, which are (i) Hoivletf s Gorge; (2) Little (or Green) Lake; (3) Green Pond; (4) Rock Gorge. (1) The Hewlett" s Gorge Station. In March, 1866, Mr. Lewis Foote, of Detroit, Mich., found the fern growing plentifully in a deep ravine of Butternut creek, five miles southeast of Syracuse, upon the line of the Syracuse & Binghamton Railway, f This ravine is commonly known as Howlett's Gorge. It is deep and rocky, especially rough on the left or northern side where the Hart's-tongue grows, though more open on the opposite side. The fern, well shaded, once grew here very plentifully, but it has been largely rooted out. (2) Little (or Green] Lake. In September of the same year, Mr. J. A. Paine visited the locality of Jamesville and extended the known range nearly to its present limits. He detected the fern on * Asa Gray in Am. Journ. Sci. & Arts, II. 41:417. 1866. I have been unable to find any earlier reference to Cooper's discovery. t Asa Gray, in Am. Journ. Sci. & Arts, II. 41 : 41?- l866 - 34 the shaded talus of cliffs which nearly surround Little Lake. Little Lake is situated approximately a mile south of Hewlett's Gorge, and a mile west of Jamesville. Paine aptly describes it as "a deep depression in the surface, walled in on all sides but one with rocks at least 100 feet high, and one-fourth of a mile across from side to side." * The open side is the eastern, t The plant was formerly very abundant on the talus at the south of the lake, but scarcely a half dozen plants may be found now, owing to the greed of picnick- ers. From thirty to fifty rods to the north of the lake the plant grows thriftily in at least three different places along the sides of two wooded ravines which occur together. (3) Green Pond. Continuing his search, Mr. Paine gave at- tention to the other pit-hole lakes of the vicinity, and found Phyllitis growing at Green Pond. White Lake and Green Pond lie near each other, a mile and a half east of Jamesville, at the base of a ledge of limestone from 100 to 200 feet high. This ledge is a con- tinuation eastward of the steep escarpment which forms the south- ern cliff of Rock Gorge, lying about a mile northwest of Little Lake. As stated, it extends eastward, and transecting the north- erly-trending Butternut valley, runs a half dozen miles farther, inci- dentally giving rise, at a given point along its base, to Green Pond. Green Pond (also called Scolopendrium Lake) is similar to Little Lake in lying like a sheltered harbor far within the irregular outline of the surrounding cliffs. It is, however, at least a third of a mile broad. The banks are exceedingly rough and strewn with fragments broken from the towering limestone cliffs. % The cliffs have extensive tali, and it is the continuous steep talus of the great U-shaped cliff which forms the shore of the lake. The fern grows pretty well up on the sides, among the fragments of Corniferous limestone, on both sides at the base of the U. The plants from the cleared (eastern) portion have mostly become of small size and winter-kill badly, owing doubtless to the comparatively recent removal of the forest, which occasions a lack of protection in win- ter and summer alike. On the western slope, as yet wooded, the plants grow to good size. * Amer. Journ. Sci. & Arts, II. 42: 281. 1866. fFor a complete description of this remarkable lake and vicinity, see the article by Prof. E. C. Quereau, entitled Topog. and Hist. Jamesville Lake, N.Y.. 111 Bull. Geolog. Soc. Am. 9: 173-182. 1898. t See article by the author, Fern Bull. 7:1. 1899. 35 (4) Rock Gorge Station. Rock Gorge, lying about a mile northwest of Little Lake, is one of the more northerly transverse valleys connecting the Onondaga and Butternut valleys. It runs east and west, trending slightly to the southeast, and is utilized by the Syracuse & Binghamton Railway. At a point about " midway its length, and on the south side of the gorge, the wall is cut back in the form of an ampitheatre which is semicircular in outline and about 125 feet deep by 250 feet wide. The walls are nearly per- pendicular, with their bases concealed by recent talus accumula- tion." * In this recess, about 40 feet from the top of the cliff, and among the loose fragments, grow about 125 extremely fine plants of nyllitis. The slope is rather steep, but the plants grow thriftily in the scattering second growth of maple and basswood, shaded by the cliff wall, which serves also as a considerable protection in winter. ^ A small number of plants have also been observed recently ( in May, 1899) to grow in a small depression some 40 rods to the west- ward and back from the amphitheatre, by Mr. Homer D. House. Mr. Paine, at the time of his vtsit to Jamesville in 1866, gave a great deal of attention to the general contour of the locality. He remarks f that "these 'highlands' before they were cleared and burned over, formed the very kind of locality where our rare fern delights to dwell, possessing all the conditions of loose limestones, rich mould, moisture and shade ; and no doubt their rocky steeps formerly abounded with it. This presumption is confirmed by the fact that on a particular part of the range, where the fire and clearing ceased and the undisturbed forest began, just there was Scolopendrium found growing in its greatest luxuriance and scattered along the bank for a fourth of a mile or so, as far as covered by rocks." d. The Perryville Station. An additional central New York station for the Hart's-tongue was discovered in July, 1898, at Perryville Falls, Perryville, by Miss Murray Ledyard, of Cazenovia, N. Y. A small stream, the Canaseraga creek, here falls fully a hundred feet, near the quarries at the railway station, and runs helter-skelter through the narrow wooded ravine below. As at *See Prof. Quereau's paper previously mentioned. He finds conclusive that this amphitheatre was once a waterfall, t Amer. Journ. Sci. & Arts, II. 42: 281. 1866. -36- Chittenango Falls, the fern grows only upon the western side of the gorge; and the two stations are otherwise very similar. Mrs. James R. Parsons, one of the discovering part}', thus writes of its occur- rence :* " The plants were * *growing in a partial open- ing among the maples, basswoods and beeches on a steep slope covered with fragments of limestone, some 30 or 40 feet from the base of the cliff. We must have found anywhere from 20 to 30 plants within a radius of as many feet." The fact of its discovery here, in a favorite botanizing field, may indicate a recent origin of this particular station. It may have arisen from the Chittenango station, which is less than three miles distant. In general, regarding its central New York distribution : It is extremely likely that the fern has by the natural clearing of the country been in some measure exterminated ; but it is a fact, nev- ertheless, that it is only in the more rugged situations of the un- cleared land that it usually grows. It stands rather as a remaining type of boreal vegetation, persisting only in such places as are well suited to it. It prefers rough, shaded tali in broken country, where extreme drought can never affect it ; where it is subjected to a uni- formly cool temperature, and protected also by considerable shade. In such a situation the fern now thrives, and doubtless will so con- tinue unless . rooted out by reckless collectors. It ought even to become settled in many additional stations in the general locality. CANADIAN STATIONS. (a) The Owen Sound Locality. Owen Sound is a port on the (Georgian Bay, the great eastern arm of Lake Huron. The town is nearly on the lake level, the rise being perhaps fifty feet to the mile ; while both east and west are cliffs Clinton on the west side, Medina on the east which form the sides of the valley. The rock is covered with soil of varying shallowness, and forested with maple, spruce, hemlock, and birch. The country all about is very rocky, and doubtless contains more stations than mentioned here. The first discovery of the fern at Owen Sound was by Prof. William Hinks, of Toronto, in 1857. He found it growing plenti- fully around the falls of a stream emptying into the Sound, t This stream is the Sydenham river, and the falls, which are situated over two miles south of the town, are known as ' ' Sydenham ' ' or * Fern Bull. 4: 74. iSgS. t J. A. Paine, in Am. Journ. Sci. & Arts, II. 42: 2Si. iS66. 37 " Inglis" Falls. Below the falls occur the plants of Phyllitis, in a " fairly heavy \vood in the valley and on the sloping sides of the deep chasm through which the Sydenham runs after its fall. The chasm is about a fourth of a mile wide, or less, its sides strewn with boulders and fragments of limestone partially buried in debris. Here the individuals were seldom large, i. e. seldom over 8 inches long.* The fern is reckoned "abundant on limestone debris under cliffs at Sydenham Falls and other localities around Owen Sound (Mrs. Roy)." t It has also been found close by "growing in a maple wood, where the rock is c lose to the surface and shows cracks of width varying from two inches to two feet. It grows [here] generally in the cracks, but also on the level ground, doubtless always where the soil is shallow. Companion plants are Dryopteris marginalis, I'olystichum lonchitis, Camptosorus, and Aspleniiini viride. % Phyllitis occurs also in a wild situation some twelve miles to the northwest of the village. Here, too, the soil is not deep, and the fern grows rather thriftily in the dark, moist, rocky woods, where the limestone ( Clinton ) comes close to the surface. In such places it grows on small hummocks slightly raised above, and so drier than the surroundings. -' The finest plants are found at a point about one mile northwest of the town, upon the loose limestone debris fallen from a bluff 40 feet high. The soil is very light, porous, and rather dry, and can- not for any length of time retain ijjoisture. But the situation seems especially favorable, and fronds grow from eighteen to twenty- three inches long. :; ~ 1'hyllitis grows about Owen Sound very much as in New York, in loose limestone debris, and additionally, in narrow limestone crevices and raised on hummocks in moist, rocky, upland woods the sort of situations especially claimed for it formerly in central New York by Paine. Mr. Jenkins obtains the best developed specimens from the station a mile northwest of the town, where the soil is the driest and most porous of any of the stations. Fur- * From correspondence, Prof. W. II. Jenkins, Owen Sound, Ontario, Canada. tMacoun, Cat. Can. PI. Part V., 268. 1890. % From correspondence with Mr. W. K. Saunders, I,ondon, Ontario. 200775 Iher, he states that the summer of 1897 being a "wet one, the t'ronds were somewhat shorter and narrower than in previous years." The habitat of the fern is, in each case, cool, well shaded, and invariably upon limestone. But Mr. Jenkins remarks further that while each is "moist so far as atmosphere is concerned," it appears that " moisture as a soil constituant is not of prime import- ance to large growth." (b) The Durham Station. The fern was found in 1883, "on Guelph dolomites, Little Sau river, at Durham, Gray county, On- tario," by Dr. H. M. Ami, of the Canadian Geological Survey/"' who has written me as follows: "The specimens obtained were growing in the narrow crevices of the cream-colored dolomites of the Gnelph formation (Silurian). They appeared to be somewhat depauperate forms, still sufficiently alive and vigorous to warrant the expectation that they would survive under the existing environ- ment. There was very little earth where they were growing, the rock everywhere being practically bare or destitute of earth or drift." Durham is about 20 miles south of Owen Sound in a very wild country, little explored botanically. Guelph dolomite is a good half magnesium carbonate. , (c) The Col ling wood Station. 1111898. Prof. Jenkins wrote me of the supposed occurrence of Phyllitis at Collingwood, Ontario. Lately I have learned through Mrs. E. G. Britton of its rediscovery by Mr. B. B. Osier, of Toronto, a member of this Chapter. Mr. Osier has kindly communicated tnost of the following data, which are set down largely verbatim : The location is lot XI. in the 3d concession of. the township of Collingwood, Grey county, Ontario. It is seven miles in a westerly direction from the town of Colling- wood, which is on the Georgian Bay, and about 50 miles east of Owen Sound. The land is about 1500 feet above sea level and of the Upper Silurian formation, a limestone of sufficient purity to be burned in neighboring kilns. The forest is ordinary Ontario growth of hard maple, mountain maple, basswood, elm, ash, beech and iron-wood, with more or less cedar and butternut. The soil is a rich clay loam, with a great deal of leaf mould. The rock on which the ferns grow is full of seams and crevices, which, together with the absence of quick evaporation (due to the dense shade), generally gives ample moisture. The region is essentially a plat- * Macouii, Cat. Can. PI. Part V. 268. 1890. 39 eau, cut into by several streams which have made valleys and gorges trending mostly to the east. The most northerly of these is the Silver. The Silver rises in a group of springs flowing perhaps 20,000 gallons per hour. The ground is moist, the shade dense, and the ground largely boulder limestone. The ferns are scattered over approximately two acres immediately surrounding the springs. The point of first discovery lies some 120 rods to the northeast of tli is point, upon the rocky tali and slopes above Kennedy creek. A few plants occur also near the banks of the Silver, about a mile from its source. The valley of the Pretty lies about two miles to the south of the Silver. The fern occurs here in some abundance along the rocky slopes of the valley, though the land is higher and drier, and the shade not so dense. Many granite and schist bould- ers occur here, but do not carry Hart's-tongue, the fern being always rooted in the limestone crevices. Mr. Os-ler is of the opinion that the fern is also to be found in the valley of the Pine and the Mad, the formation and aspect being similar. The Holly fern always occurs with the Hart's-tongue at these stations, and near the springs the Walking-leaf in consider- able abundance. Mr. Osier has kindly furnished a series of photo- graphs of the ferns in their native environment. In a later letter, attention is called to the fact of the fern's destruction in quantity by young cattle. (d) The Woodstock Station. The Hart's-tongue was discov- ered near Woodstock, N. B., in the late fall of 1882 by James Sut- ton, a gardener in the employ of Mrs. Charles Connell, of Wood- stock. During a visit to this village the next September, the atten- tion of the late Peter Jack, of Halifax, N. S., was directed to the plant, a single rather undersized specimen, which had been pre- served in Mrs. Connell's greenhouse. Mr. Jack immediately rec- ognized it as " Scolopendrium vulgare." * From Mrs. Dibblee, nee Connell, of Woodstock, and Mr. G. U. Hay, of St. John, N. B., I have learned that it was collected some six miles to the westward of Woodstock, upon the "Richmond Road," near the Meduxnakik river where it was open to the northwest. Mrs. Dibblee has stated also that the station ' ' has been all burnt over, ploughed up, and is now a fine farm," and that the fern grew upon what is commonly called * See Notice of New and Rare Plants by George I,awson, in Proc. and Trans. Nov. Scot. lust. Nat. Sci. Q. 71-72. 1883-6 ; also a paper by G. U. Hay, entitled Potai4y of the I'pper St. Joint, in Bull. Nat. Hist. Soc. N. Br. No. 2, pp. 31 and 37. 4o there the shale land which supports a good growth of elms, butter- nuts and ash. Mr. Jack visited the station at the time, but without finding further plants. After his return to Halifax, the gardener upon dili- gent search secured a number of additional plants, four of which were forwarded to Mr. Jack, who later presented one to the Nat- ural History Society of New Brunswick, and a frond to the Nova Scotia Institute of Natural Sciences. The Eaton Herbarium con- tains also two fronds of the var. marginatum, collected by Sutton in July, 1885. Mrs. Dibblee states that the fern was brought in several times from the same place. Mr. Hay adds that quite a number of them still thrive in the conservatory and upon the rock- ery at her home, but that plants presented to him at various times have not survived. Mr. John Macoun has (in correspondence) suggested the pos- sibility of the fern's having become first established as an escape, but a thorough knowledge of the facts attending its discovery has convinced both Mr. Hay and myself that it was undoubtedly native. In fact, Mr. Sutton has stated that he once found the fern in a ravine farther in the woods, about eight miles from town, but that he has not again seen it there. Mr. Hay has long intended a sys- tematic search for the fern about Woodstock. Such a search would probably result in its discovery somewhere in the general vicinity. THE TENNESSEE STATIONS. * The Hart's-tongue has been found in two localities in Tenn- essee, viz., near Post Oak Springs, and at South Pittsburg, of which the latter only has been known in literature. a. The Post Oak Springs Station. In 1849 ^ r - A.Gattinger de- tected the fern a short distance west of the village of Post Oak Springs, in Roane county. About one mile southwest of the vil- lage occurs a pool in an open cave in the front ( southeastern ) edge of the hill, and from this issues a small stream to the eastward. Over the top of the hill, i. e., on the northern side, and distant about a half inile, occurs another open cave, called the " dry cave," about the mouth of which a few plants were found by Dr. Gattinger. This cave is about one mile directly west of the village. * In his Tennessee Flora ( 1887) p. 102, Dr. Gattinger refers the fern to " New Pittsburg," a mistake for South Pittsburg, and adds "not found [there] by myself." The previous station (Post Oak Springs) concerning which he has lately written me, was unfortunately overlooked. According to the Kingston Folio of the U. S. Geological Sur- vey Atlas, the immediate rock formation is the Knox Dolomite, or Magnesium Limestone of the Lower Silurian. But Dr. Gattinger avers that the Chattanooga Black Shale (Devonian) and the Fort Payne Chert (Lower Carboniferous) both crop out here. The fern probably occurs in the latter. * Immediately above is the Bangor Limestone, which supports the fern at South Pittsburg. b. The Smith Pittsburg Station. The Hart's-tongue was discov- ered growing in a deep sink-hole near South Pittsburg, by Major Cheathem, in 1879. t South Pittsburg is a town on the Tennessee river, about three miles north of the Alabama boundary. Some two miles southwest of the town, two spurs of the Cumberland mountains, extending southeast into the level plain of the river, form a narrow valley or "cove," as they say in Tennessee. To reach the sink-hole, follow the cove a half mile, or until half way up the mountain. Sixty feet to the left of this narrow valley and about sixty feet above, there is an irregular fissnre in the Mountain lime- stone, sixty feet long by twenty to forty feet wide and ninety-two feet deep. Upon examination, a good-sized spring is found to issue from a cave not more than twenty yards farther up the hill. This spring, tumbling perpendicularly into the hole, strikes a pro- jecting ledge some forty feet below. The water splashing from tth ledge has worn a deep depression in the opposite side, and it is chiefly upon this slope that the fern grows. The area covered by the ferns is not over 200 square feet, and contained (in 1898) about 1 10 mature plants. A few are variously distributed along the sides and edges of the chasm, with Asplenium parvulum. The soil is a sticky, light-colored clay, formed from the disintegrated shales of the upper mountain. The ferns are found mostly about fifty feet below the surface, and are so sheltered that some of them the direct sunlight never reaches, and can possibly reach none of them longer than two hours a day. There is little variation in tempera- ture, naturally, and they are always dampened by the spray of the falling water. After its first drop of forty feet, the water trickles in small streamlets over the rock walls and fragments the remaining fifty-two feet, to disappear in a narrow fissure at the bottom of the * Dr. Gattinger has kindly furnished me full particulars regarding this sta- tion. I am under obligation also to Mr. David White, of the U. S. Geological Survey, Washington, D. C., for information relating to the geology. fBull. Torr. Bot. Club, Q: 350. 1879. 42 pit. It reappears at the surface nearly two miles distant, on the level flat a short distance from the river. Here it is called the Blue Spring.* As nearly as can be ascertained without a special visit to the place, the formation in which the sink-hole occurs is the Bangor "rotten" or "Mountain" limestone. [See note III in thesis by the writer, previously mentioned. ) This limestone includes a few- beds of shale, several occurring near the top, so that the presence of the clayey habitat is easily explained, t SEVERAL SUPPOSED STATIONS. (a) Chiapas, Mexico. In his Ferns of North America, J Prof. Eaton cites the Hart's-tongue as occurring at "Chiapas, Mexico." The specimens (Eaton Herbarium) upon which this determination was apparently based, were collected in ' ' cool regions ; crevices of rocks in the bottoms of caves (296-307)" at Chiapas, Mexico, by Ghiesbreght. || They are not to be referred to Phyllitis Scolopen- dnum, but rather to Phyllitis Linden! (Hook. ) a species clearly distinct from the former. The habitats of Ghiesbreght' s and Lin- den's specimens are rather diverse, but not more so than those of many species of Polypodium of the same region. (b) Sitka, Alaska. Milde^j records the fern from " Insula Sit- cha," and adds "(herb. caes. Petrop. horti bot. )," indicating that the specimens are preserved in the St. Petersburg Herbarium. The *The greater portion of the foregoing has been kindly contributed by Mr. nd Mrs. Joseph H. Lodge, of South Pittsburg, who at the time (1898) of their ivestigation forwarded several living plants to me. I have described the sta- on in some detail, since an article has recently appeared (James II. Ferriss, Bull. 7: 98. 1899) indicating doubt as to the existence of the fern in that tation at present. It is evident that Mr. Kerriss missed the sink-hole in question, which is not remarkable, since there are several in the near vicinity. tSafford and Killebrew, Elem. Geol. Teiin. p. 153. tD. C. Eaton, Ferns N. Am. \: 247. 1879. I See Cat. Coll. Ferns So. Mexico, mainly at Chiapas, by A. Ghiesbreght, 1864-70. This pamphlet (pp. 10), kindly loaned me by Prof. Underwood, is evidently a reprint. The determination of the ferns was accomplished by Hall, probably under Prof. Eaton's supervision. The two sheets have been kindly loaned me by Prof. A. W. Evans, of Yale University. \ScolopendriumLindeni Hook., well figured and described (Hooker, Ic. PI. II. \:pl. 488. 1842) from specimens collected "on old oaks, Chamulars, Prov. Chiapas, Mexico," by Linden, \\. 1543. , Fil. Europ. et Atlant. 90. 1867. 43 Hart's-tongue is not uncommon in Japan, and may, like some other species (notably Dryopteris montana, for a long time sup- posed not to occur in North America), have an eastern extension across into Alaska, and down the Pacific coast into British Colum- bia. An examination of the material in the St. Petersburg Herbar- ium, would of course establish the identity of the specimen, but would not prove the authenticity of this station, since it has fre- quently happened that plants from Russian territory upon both sides of Behring Sea have been carelessly and indiscriminately labeled. (c) I'ancoui'er Island, British Columbia. In the fall of 1898, Mr. Hamburg, a Swedish botanist, then just returned from a col- lecting trip in the West, assured me that he had recently collected specimens of this species growing upon Vancouver Island. The statement was, unfortunately, not substantiated by specimens. (d) Maniioulin Island, Canada. Professor W. H. Jenkins has written me of the reported occurrence of the fern upon this island in Lake Huron, rather more than a hundred miles northwest of Owen Sound. I have found no further reference to its occurrence here. (e) Louisville, Kentucky. John Williamson * was inclined to discredit the reported station near Louz'sville, mentioned by Dr. McMurtrie. t A reference to Dr. McMurtrie's book (Library of Congress) shows the following entry : " Asplenium Scolopendrium Crt. Hart's-tongue," along with a few other ferns reported from the vicinity. It may be taken for granted that the Hart's-tongue was not confused with another species, as the chances for such an error are indeed small. The fact that Williamson was unable to find the fern might indicate either that it has disappeared from the vicinity, that -'vicinity" was made to include a considerable terri- tory about Louisville, or that the fern has since been overlooked near the city. The second supposition is the most likely. It seems that Williamson and later botanists would have found the fern if it still occurs in the immediate vicinity of Louisville ; and it is not probable that it has been exterminated. At the time (1819) * Williamson, Ferns of Kentucky, p. iv. 1878. tH. McMurtrie, M. D., Sketches of Louisville and its Environs, Ed. i, p. 229 Louisvile, 1819. This book includes a " Florula Louisvillensis, or a catalogue of nearly 400 genera and 600 species of Plants, that grow in the vicinity of the town, exhibiting their Generic, Specific, and Vulgar English names." 44 McMurtrie's book appeared, the Geddes, N. Y. station, was the only one then known in America, having been recorded by Pursh just five years previously. CONCLUSIONS. I have described with considerable fullness the American local- ities where Phyllitis Scolopendrium is known to have been found. The striking and invariable characteristic of its environment ap- pears to be an affinity for limestone rocks the Corniferous in cen- tral New York ; the Guelph Dolomite and Clinton in Ontario ; the Bangor (Mountain) limestone in Tennessee, representing consid- erable range in choice of formations. The question naturally arises : Why should the number of known localities be so small ? As a matter of fact, the Hart's-tongue is, as I have indicated, far more common than has usually been thought ; and it will prob- ably turn up continually in some of the less explored regions of Canada, especially in the northwest. It appears to require for its best growth a cool, well-shaded limestone ravine, talus, or sloping woodland, with rich wood soil, for the most part sufficiently porous to allow free drainage, but firm enough to retain considerable moisture. Such conditions occur in hundreds of glens in the United States and Canada. There is moreover, quite a wide vari- ance, in the wet heavy clay of the South Pittsburg sink-hole, from the rich porous soils and loose leaf moulds of the central New York stations, or the scant soil, of the limestone crevices in some of the Owen Sound stations. The South Pittsburg clayey habitat is, to be sure, somewhat anomalous, but it appears that almost any soil upon limestone will support the fern under the right temperature conditions, perhaps the most important factor now operating for or against the fern's survival and in determining its future distribu- tion, is the presence of a constant low temperature. The James- ville pit-lakes have a uniformly cool temperature from day to day. The Chittenango gorge is deep and the fern is well shaded. Ferns growing from fifty to seventy-five feet below the surface, as in the irregular South Pittsburg chasm, cooled by a waterfall, can be affected only in slight degree by extremes of temperature. The Canadian stations are mostly near streams. It is true also that extremes of cold do not seem to affect this fern deleteriously. It remains evergreen through ordinarily severe winters, except when unduly exposed by removal of protective forest growth. And so, while the fact of an even low temperature does not adequately 45 explain the causes resulting in the present peculiar distribution, I think it does throw considerable light upon the relationship between environment and present distribution. Since arriving at this conclusion, I have chanced to note that in several British works, considerable attention is paid to the fact that the Hart's-tongue occurs regularly " in caves, on the seashore and in other cold and damp situations, " and again, "more especially about the mouths of caves, deserted mines, at the borders of wells, where there is a current of cold or moist air." It is, of course, a well known fact that plants characteristic of high northern latitudes, are found in more or less abundance upon mountain peaks of the more temperate regions thousands of miles to the south. The accepted explanation is : that during the glacial epoch the plants, gradually forced south by the advancing ice sheet, upon the northward retreat of the ice, moved up the mountains, seeking to maintain accustomed environmental conditions, * or, for the same reason, advanced to the northward. Thus, the Hart's- tongue occurs in America mostly in the north. It has, as I have remarked, been usually regarded as a boreal type. I believe that it was once far more common than at present, and that it will fre- quently be found in the north, perhaps, as I have suggested, stretching across from Asia to Alaska, and down the Pacific coast. In the United States it has yet to be seen west of the Mississippi, though it may possibly occur along the upper tier of States, assum- ing that it follow a belt parallel to the lower limit of glaciation. Especial search and exploration in favorable localities would not be without good results in general, and would very likely result in further extensions of range for what has commonly been regarded as one of the rarer American species of ferns. The distribution of the Hart's-tongue in Great Britain is pecu- liar as well, and has been commented upon by Mr. Druery (Choice British Ferns, p. 14. 1888) at some length. He remarks that its comparative rarity in Scotland is the more unexpected in view of the " innumerable glens which abound there and seem a very beau ideal of a habitat for it;" and adds, moreover, "this fern is one of the * All especially interesting exposition of the facts and causes having to do with the distribution of species (especially North American), is contained in the latter portion of Dr. Asa Gray's " Memoir on the Botany of Japan, etc," in Mem. Am. Acad. Arts and Sci. II. &' l8 59- - 4 6- least dainty in its requirements, seeming to have no antipathy in the matter of soil or position, and in many localities thriving in abundance under the most adverse conditions." Mr. Druery is inclined to the belief that this paucity in certain sections is due to the "greater or less predominance of certain forms of minute insect life. * * Just as we find in our gardens that certain vermin attack and destroy certain plants, so it is only reasonable to as- sume that either the spores or prothalli of these ferns are the favo- rite food of some of the minuter insects, in order to explain the absence of adult plants. Climatal conditions are, of course, a potent factor, but do not account for all the phenomena observed." Mr. Druery then cites the cases of certain exotic ferns which attract slugs or snails, and must be grown in isolation. Such a one is Cainptosorus, in England. "Other plants," he writes, "are espe- cially subject to the attack of wood lice ; and so, doubtless, such special appetites characterize also the minuter and microscopic insect world, and as it is manifest, when we consider the myriads of spores which are shed in vain, that these must become mainly the food of such tiny creatures, we need hardly seek further for a solution of the mystery. A harder or softer envelope to the spore, or a more or less attractive flavor in this plant itself, would deter- mine for or against its survival in the struggle for existence." It seems necessary to suppose that some such cause operates to effect the odd distribution both in England and America. In March, 1898, I noticed the presence of one of the Lace bugs (Tingitidae), upon plants at Jamesville, in considerable numbers. They seemed mostly to attack the spores, but were found appar- ently destroying the leaves as well. Snails are often found depend- ing from badly eaten fronds. These scant observations tend to substantiate Mr. Druery' s proposition. It is indeed perfectly sup- posable that in certain of its various stages of development, it may peculiarly attract vermin to itself, which so greatly impair its vitality as to lessen its chances for survival and reproduction. A series of careful observations and experiments along this line would be of the greatest interest. U. S. National Museum, \Yashington, D. C. UNIVERSITY OF CALIFORNIA LIBRARY Los Angeles This book is DUE on the last date stamped below. MAR 3 11965 Form L9-25m-7,'63(D8618s8)444 AA 000876428 JNIVERSITY of CALIFORN1 LIBRARY 1-