Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 48, 1915
This text is also available in PDF
(5 MB) Opens in new window
– 253 –

Art XXX.—Studies in the New Zealand Species of the Genus Lycopodium: Part I.

[Read before the Philosophical Institute of Canterbury, 3rd November, 1915.]

Plates XVII, XVIII.

In these studies I hope to put together the results of my observations on the life-history of the New Zealand species of Lycopodium. In several of the papers which have been published on the subject of the genus Lycopodium the paucity of information available for the purpose of a comparative study has been commented upon. The present study seeks to bring into the arena of Lycopodium problems additional data with regard to the occurrence and habit of the mature plant, the occurrence and structure of the prothallus, the nature of the dependence of the young plant upon the prothallus, and the vascular anatomy of both the “seedling” and the full-grown plant.

The chief modern writers on the subject are Treub, Goebel, Bruchmann (6), Lang (14), and Thomas (16) on the prothallus; Jones (13) on the stem-anatomy; and Sykes (15) on the morphology of the sporangium-bearing organs. Bower (4) has thoroughly analysed and co-ordinated all known Lycopodium facts in connection with his well-known theory of the origin of the sporophyte: he concludes that the genus is to be read as a progression from a simple Selago-like form. Lang (14) has also given a comparative analysis of the facts, and, following Treub, has arrived at conclusions in which the species which belong to the L. cernuum cycle of affinity are viewed as the most primitive in the genus, the whole genus being read as a reduction series. Thomas and Sykes support this second interpretation, but Goebel doubts the primitive nature of the protocorm, on which, of course, this view is largely based. Lady Isabel Browne (5) has lately proposed a third interpretation, according to which the protocorm is to be regarded as a reduced stem. Bruchmann's interpretation of the various types of Lycopodium prothallus would separate so widely the different sections of the genus from one another as practically to deny that they are interrelated at all New Zealand contributions on the subject, in addition to Thomas's preliminary note on the prothallus of Phylloglossum (16), have been Miss Edgerley's account (8) of three species of prothallus, and my own papers on Lycopodium stemanatomy (11) and on the protocorm (12). Unfortunately, I have not had the advantage of consulting, for the purpose of this present study, either Treub's or Goebel's original publications. My knowledge of the work of the former has been derived from various references to and figures from it in Bower's “Origin of a Land Flora” and in other standard books of reference, and also from Treub's own preliminary note in the “Annals of Botany” (17), and Bower's “Review of L. Phlegmaria” (2).

After this paper had been put into the printer's hands I discovered the prothallus and young plants of the endemic L. ramulosum. Through the courtesy of the Editors I have been able to add an account of these, with several figures.

– 254 –

I desire to express my gratitude to Dr. L. Cockayne, F.R.S, and to Dr. C. Chilton, C.M.Z.S., Professor of Biology in Canterbury College, Christchurch, for the encouragement and advice they have given me during the last two or three years, and to the former also for many suggestions with respect to this paper; also to Mr. C. E. Foweraker, M.A., of the Biological Laboratory, Canterbury College, who has given me great assistance in the preparation of objects for the paraffin bath.

I. Occurrence and Habit.

L. Selago Linn.

This species is common in Europe, and is well known from the investigations of H Bruchmann (6) and C. E. Jones (13) and others, so that it will need but slight notice in this paper. It occurs commonly throughout the South Island of New Zealand in damp places as a member of fell-field, herb-field, and sometimes subalpine Nothofagus forest, and elsewhere on the high mountains.

L. Billardieri Spring.

This species is found in the forest throughout New Zealand. Though typically an epiphyte, growing pendulous from tree-fern trunks and the upper branches of forest-trees, it also not uncommonly grows on the ground. In North Auckland it is met with on peaty soil in groves of Leptospermum. On the volcanic islet of Rangitoto, Auckland Harbour, it grows on patches of humus among the blocks of scoria. In Southland and Stewart Island it occurs frequently on the damp forest-floor on patches of humus at the base of large tiees. In one instance I found a single “seedling” plant growing on a bush-road clay-cutting. The individual plant consists of a main stem, which in its short underground region is sparingly branched. This underground portion is covered with scale leaves, and bears a number of roots at its lower end. These [ unclear: ] oots are much branched, and are covered in their terminal portions with a mat of rhizoids. At or near the surface of the soil the stem branches dichotomously several times to form a tuft of aerial stems. When the plant is epiphytic these aerial stems hang in tresses from 1 ft to 4 ft. in length (see Cockayne, L, Fig 6, in “Report on a Botanical Survey of the Waipoua Kauri Forest,” 1908), when terrestrial the plant is more or less upright, and in some instances it is then hardly to be separated from L varium Several plants generally grow together, their subterranean portions thickly interpenetrating the patch of humus. The appearance of the plant, with its dichotomously branched aerral stems and numerous long terminal tetragonal fertile spikes, is well illustrated by Pritzel's figure of L Phlegmaria in Engler and Prantl (9).

L. varium R. Br.

This plant is stated by Cheeseman (7) to be probably only an extreme form of L Billardieri. It is sometimes epiphytic, but it occurs in those parts of New Zealand subject to a semi-subantarctic climate, and on Stewart Island and the Subantarctic Islands of New Zealand more commonly as an erect, stout, rigid terrestrial plant. On Stewart Island I found it growing on the forest-floor in large patches, some of which were as much as 12 ft. across.

– 255 –

L. Drummondii Spring.

There is only one locality, so far as is known, where this species occurs—viz., the Sphagnum bog at the east end of the small Lake Tongonge, at Kaitaia, North Auckland. I desire to thank Messrs. H. Carse and R. H. Matthews, of that neighbourhood, for kindly conducting me to the spot in January, 1914. In a recent letter the former states that the lake and bog are to be drained, so that probably this species, so rare in New Zealand, will disappear. The main stem of the plant is never more than 6 in. in length, and is generally branched several times. It creeps above ground, but is tightly bound down to the mossy surface by the adventitious roots, one of which is borne at the junction of each branch with the main stem. The cones are borne singly on erect peduncles, and stand from 2 in. to 4 in. in height. In several instances a fertile region was seen to be divided into two lengths by the interposition of a short sterile region, and in other cases the old cone of the previous year was observed to have grown on to form the new one.

L. laterale R. Br.

I have studied this species more especially on the clay “gum-land” in boggy localities around Kaitaia and on the Auckland Isthmus. In the latter locality it occurs at the margins of the small bogs which occupy the numerous hollows among the clay hills, growing amongst certain Cyperaceae and Gleichenia dicarpa var hecistophylla. Around Kaitaia and on the Peria Gumfields it grows extremely abundantly on the open damp hillsides. The surface soil of these gum-lands consists of a peaty humus, which for the greater part of the year holds much water, but which during the summer months is generally more or less dry. The adult plant consists of an irregular and much-branched colourless rhizome, which ramifies through the soil in all directions. The shorter branches emerge at the surface to form the e [ unclear: ] ect aerial shoots. These latter, when growing amongst thick scrubby vegetation, are extremely slender, and attain a height of from 2 ft. to 3 ft. On open ground they are short and stout, and often reddish in colour. The cones are short, and are normally lateral and sessile. In some cases, however, individual cones are borne on short leaf-covered peduncles, and they must then be regarded as terminal. Pritzel's description of this species in Engler and Prantl is rather misleading. He there states that the Cernua section comprises forms without a widely creeping main axis, mostly like a little tree. It is to be noted that it is only the aerial branches which are tree-like in the two New Zealand species L. laterale and L. cernuum. The main body of the plant in L laterale is subterranean and widely ramifying.

L. cernuum Linn.

This species is well known from the writings of Treub and of Jones (13). It grows very abundantly throughout the northern part of the North Island of New Zealand on clay moorlands as described for the preceding species. It thrives especially in North Auckland amongst scrub vegetation of the Gleichenia-Leptospermum association, individual plants often attaining to a length of 12–15 ft., and the upright branches to a height of 1–4 ft. It is also extremely common on the Volcanic Plateau, in the neighbourhood of hot water and near fumaroles. The

– 256 –

main stem of the adult plant is above ground, and has a serpentine habit of growth. It extends in a succession of loops and nodes, at each node the stem being fastened to the ground by a group of adventitious roots, which arise in the first place immediately behind the stem-apex on its ventral side, while from the loops arise the branches of limited and unlimited growth. The branches are borne laterally right and left on the main stem, but the erect tree-like fertile branches take their origin from its dorsal side. Here, again, Pritzel's description is misleading: only certain of the branches—namely, the fertile ones—are tree-like; the main body of the plant is widely creeping and branched. The figure which he gives in Engler and Prantl illustrating this species is that of an erect fertile branch only; the group of roots at the base should not be so figured. There are numerous short cones borne at the ends of the branchlets in the fertile regions.

L. densum Labill.

This species grows abundantly throughout the Auckland Province on clay land amongst light open “scrub” vegetation, and also in the more open parts of kauri forest near its outskirts. (For general habit see Cockayne, L., loc. cit., Fig. 16.). The main rhizome, which is from 4 ft. to 10 ft in length, is stout, and there are also subterranean branches both of limited and also unlimited growth, borne laterally on the main stem. The branches of limited growth emerge from the ground as rigid, erect, much-branched, tree-like, aerial shoots, generally from 1 ft. to 3 ft in height. I have often observed that when the plants are growing amongst tangled scrub the aerial shoots may be as much as 8 ft or 9 ft. in height, and in some cases remain totally unbranched. Stout adventitious roots arise ventrally from the main rhizome and branches, and are borne singly, generally immediately behind a point of branching. There are three distinct varieties of this species, corresponding to differences in the general size of the leaves with which the aerial branches are covered. The numerous short cylindrical cones are solitary and terminal on the ultimate branchlets. In the particular variety which is characterized by the acicular form of its leaves the branchlets on which the cones are borne are more or less modified as peduncles.

L. volubile Forst.

This species is common throughout New Zealand, excepting in the driest districts, growing freely amongst Leptospermum and other heath-like vegetation. It has a scrambling habit, or at times is a winding liane, spreading over the ground or over low-growing bushes Individual plants are often so much as 12–15 ft in length. There is a main axis of growth, on which are borne laterally, in the plane of the surface on which the plant is growing, branches of limited and also of unlimited growth. The leaves are dimorphous, as has been noticed and commented upon by Boodle (1), the larger laterally borne, sickle-shaped, and distichously spreading, the smaller linear, scale-like, and borne dorsally and ventrally. The distichous character is for the most part confined to the smaller branchlets on the branches of limited growth An account of the development of heterophylly in this species and in L. scariosum has been given in a former paper by the writer (11, p. 366). Here it need only be mentioned, as Goebel has already pointed out, that certain leaves on the main shoots are hook-like,

– 257 –

and so probably climbing-organs (10, p. 346). Adventitious roots are borne ventrally at intervals along the main axis. In places where the plants are scrambling over low-growing bushes these roots may attain a length of 3–5 ft. before reaching the ground. In the late winter and early spring a very characteristic feature is the thick envelope of mucilage which covers from 3 in. to 12 in. of the growing root-tip before it has reached the ground. The fertile spikes are thin and cylindrical, and from 1 in. to 4 in. in length, and occur in large terminal much-branched panicles. They are figured in Engler and Prantl (9). The fertile branches are to be found for the most part in those regions of the plant which are elevated on some low-growing vegetation, and they are thus generally pendulous. Although the panicles of spikes in their normal form are very distinct from the ordinary vegetative branches, yet close observation shows that all stages of transition may occur. Isolated sporangia are sometimes to be found on sterile branches; in other cases fertile and sterile branchlets are indiscriminately mixed; while in others, again, sterile tracts may appear in the spikes themselves.

L. ramulosum T. Kirk.

I have gathered this species from the peaty flats at the head of the Rakiahua Arm of Paterson Inlet, Stewart Island, and from bogs in the neighbourhood of Hokitika and Kumara, Westland. It occurs abundantly, covering the ground with mats of interlacing plants. It is common in such a habitat all over Stewart Island, ascending to the summits of the mountains; and is found also in bogs, both lowland and subalpine, throughout Westland and north-west Nelson. The individual plant is very short—from 2 in. to 9 in. in total length—and is irregularly branched both above and below ground. The study of this species as growing in Stewart Island shows that the subterranean portions are of two distinct kinds. Those stems nearer the surface are whitish in appearance, and are thickly covered with scale leaves; from these the aerial branches arise. Those portions which penetrate the peaty humus more deeply are brownish in colour, and are more or less naked of scale leaves, and are the ones which more frequently bear the adventitious roots. The two kinds of stem arise from one another without transitions. The aerial branches are procumbent or ascending, and are much branched. The short solitary cones are borne terminally on erect branches, but occasionally they may occupy a lateral position.

L. fastigiatum R. Br.

This species is common on open mountainous country throughout the South Island, especially on tussock grassland. It also occurs in subalpine Nothofagus cliffortioides forest, where it assumes a more mesophytic habit. In Southland and Stewart Island it also descends to low levels. There is a subterranean creeping main axis, which is usually from 1 ft. to 3 ft. in length, but which may be as much as 5 ft. long. Branches of limited and also of unlimited growth arise from the main axis, the former emerging from the surface to form the erect greenish or reddish tree-like aerial branches. These latter are from 6 in. to 12 in. in height, and are densely branched, but when growing amongst thick tussock-grass are much more slender in habit, and are taller. When growing on sour peaty soil the branches may be flattened to the ground. The cones are from 1 in. to 2 in. in length, and are borne, usually singly but sometimes two or more together, at the end of the branchlets on distinct peduncles.

– 258 –

L. scariosum Forst.

This species, like the last, occurs fairly commonly in open situations at fairly high elevations, especially throughout the South Island. In such a situation it is always creeping on the surface of the ground. Individual plants may have an extreme length of 5–6 ft., but are generally shorte [ unclear: ] . The stout and rigid main stem bears branches both of unlimited and of limited growth. The former are closely adpressed to the surface of the ground. and bear the adventitious roots. The branches of limited growth are heterophyllous, and are markedly flattened in the plane of the ground. The development of the heterophylly is different in this species from that in L. volubile, and has been fully described else-where (11, p. 366). The cones are from. 1 in. to 2 in. in length, and are borne singly at the ends of branchlets on peduncles, in the same manner as has been described for L fastigiatum, and as is so well known in the European species L clavatum. When growing amongst thick fern vegetation—as, for example, on the tailing-heaps in the neighbourhood of the old alluvial gold-mining claims in Westland—it is noticeable that this species may show an almost entire absence of the usual dorsiventral appearance of its branches. The lateral branches are erect, and, except in the older parts of the branches, the leaves are scattered and tend to be acicular in form, while in the ultimate branchlets they are reduced to mere scales. The general habit of the plants in these cases is almost scrambling, and the long, rigidly erect, naked, and closely crowded ultimate branchlets present a very characterstic and forest-like appearance. The tips of certain of these branchlets become fertile and develop as cones, while isolated fertile regions may also be found occasionally on other branchlets.

Summary

Of the foregoing eleven species of Lycopodium native in New Zealand, one is generally an epiphyte (L Billardieri), another is sometimes epiphytic, but more frequently terrestrial (L varium), five occui in more or less wet habitats (L cernuum, L laterale, L Drummondii, L. ramulosum, and L. Selago), and four in dry and at times fully exposed localities (L. volubile, L scariosum, L. fastigiatum, and L densum) All of these species except L Selago and L cernuum are confined to the countries and islands of the South Pacific Ocean, so that possibly they are not well known either to European or to American botanists. A somewhat detaled description of their occurrence and habit has therefore been given, but more exact ecological studies are demanded before a true estimate of their life-requirements can be gained. Some of these species are amongst the largest of modern Lycopodiums, as, for example, L volubile, L scariosum, L densum, and L cernuum; and also, as will be described below, the vascular cylinder of the stem in these species is greatly developed. Two show dimorphism in their leaf-structure (L volubile and L scariosum), but other characters show that these two species are not closely related. In several of the species the character of the fertile regions is variable, and also the habit of the whole plant in one or two cases varies, perhaps epharmonically, under differing conditions. Not a few of the New Zealand species occur most abundantly in localities formerly occupied by forest. For example, L densum and L laterale find their most luxuriant development on the clay gum-lands of the Auckland Province, areas occuped at no greatly distant bygone period by the kauri-tree forests; L. Billardieri and L volubile also occur freely in these localities, L cernuum is also abundant in the same locality, but

– 259 –

it shows still greater luxuriance in the neighbourhood of hot-water streams. L. scariosum and L. fastiatum frequently occur in mountain areas of both Islands, on hillsides from which the forest has retreated owing to climatic changes or to the hand of man, but the latter species is also a common plant of subalpine southern-beech forests. The two epiphytic species very readily adopt a terrestrial habit both on the forest-floor and also in more open situations in which the conditions are practically epiphytic. In Westland L. volubile and L. scariosum are very common on the heaps of tailings around abandoned alluvial goldmining claims, and the endemic L. ramulosum is abundant in the man-induced Sphagnum bogs in the same localities.

II. Occurrence of Prothalli and Young Plants.

L. Billardieri.

Prothalli and young plants associated together were found in two different localities, and young plants alone in two others. On one occasion a large number of young plants were found growing at the top of a nikau palm in the Waikumete Bush, Auckland, in a mass of humus through which the stems and roots of the adult plants were ramifying. One or two old prothalli attached to well-grown young plants were obtained in this case. On another occasion twelve prothalli were obtained in the month of January in the forest at Pipiriki, Wanganui River, from a mass of humus in the fork of a tree. Here, also, old plants were present. Of these prothalli, seven were without young plants attached, and five bore young plants in different stages of development. The prothallus of this species grows completely buried in the humus. The prothalli are easily seen, by reason of their whitish appearance, in the midst of the dark humus when the latter is dissected in water, but may be mistaken for a mass of young root-tips.*

L laterale.

Young plants and prothalli of this species were discovered by me in three different localities on the Auckland Isthmus during the summers of 1905 and 1914. In two cases they were growing on patches of peaty humus which had been overturned by gum-diggers, and occurred close to the edge of boggy ground in the immediate vicinity of adult plants. In the third case they were growing fairly abundantly on a patch of damp soil, which was sparsely covered with short moss, in the midst of a clump of mature plants. Altogether eight prothalli were found, as also were many young plants in all stages of development. Diligent search was made in several localities in the Mongonui County, North Auckland, on the open hillsides where this species grows luxuriantly, but in no case were the young plants seen. I adopted the plan of cutting out small turves of the soil in which young plants could be seen, and reserving them for examination under a dissecting microscope. The process of examination proved exceedingly tedious, for the peaty humus was closely intermixed

[Footnote] * I have found also the young plants of L Billardieri var gracile growing on the trunk of a tree-fern near Lake Kameri, Westland. The distinguishing characters of this variety are its small, slender form, almost unaltered character of the fertile leaves, and the fact that the fertile regions are not confined to the ends of the branches.

[Footnote] † This is the official spelling of the name of the county according to the Counties Act of 1908, although the correct spelling is Mangonui.

– 260 –

with a small moss, and had to be dissected in water very small portions at a time. The prothalli are almost colourless, and are very small and delicate, and hence not easy to distinguish. They were never found attached to young plants which bore more than two protophylls. The youngest plantlets are also very minute, but are more easily seen amongst the moss and soil by reason of their vivid green colour. Many of the localities where L. laterale grows become dry during the summer, and probably in some summers the young plants all die; but during spring and early summer they doubtless are growing in large numbers in favourable spots. A close examination of disturbed soil—e.g., gum-diggers' holes, hoof-prints, &c.—along the edge of the bogs around Waikumete and Henderson, on the Auckland Isthmus, at that season of the year, and in all probability generally, results in the discovery of colonies of the young plants of this interesting species.

L. cernuum.

On many occasions, and in many different parts of the Auckland Isthmus, as also in the Mongonui County, I have noticed young plants of this species growing in the vicinity of older plants. The spores would seem to germinate very freely. But, as in the case of L. laterale, a dry summer would bring about the destruction of most of the plantlets. A damp clay bank or a shaded roadside cutting in the neighbourhood of adult plants is the best place to search for the young plants and prothalli—i.e. they are more in evidence under artificial than under natural conditions. Such a clay bank, if damp, shaded, and old enough for a thin covering of moss and slime fungus to have appeared on it, invariably contains the young plants of this species, generally in great abundance. The same method of search for the prothallus was adopted as in the case of the last species. The very young plants and prothalli of L. cernuum are difficult to clean owing to the intimate penetration of the clay and slime by their numerous rhizoids L. cernuum and L laterale both grow commonly from the Auckland Isthmus northwards, but their habitats do not overlap. The former species keeps to the higher parts of the undulating clay gum-lands, while the latter is to be found in the hollows and on the damp lower parts of the hillside. I have never found the prothalli and young plants of both species growing together. However, even if this had been so, it would not have been very difficult to distinguish between them, for in spite of the similarity in appearance both of their prothalli and of the very young plantlets there are some characteristic differences.

L. densum.

Frequent and long search for the young plants of this species in various parts of the Auckland Province, where it grows abundantly, has never met with much success. On one occasion, in the neighbourhood of the Bay of Islands, I found two young plants, one of which still showed a large “foot”. The plants were growing in a grove of Leptospermum on a patch of thick damp moss. At the time when they were found it was impossible to spend more than a few minutes in searching, but I have no doubt that a closer examination of the ground would have resulted in the discovery of still other young plants. The rarity of occurrence of prothallial plants of this species may be due simply to the fact that its usual habitat is a dry open one, where damp shaded spots, favourable to the development of the prothallus, infrequently occur.

– 261 –

L. volubile.

I have been able to find young plants and prothalli of this common species in several different localities in both the North and South Islands, in some cases in abundance. Several young plants and one very large prothallus were dug out of a clay bank at Kaiaka, North Auckland. Between sixty and seventy prothalli were unearthed from an old clay cutting at the edge of the forest near Henderson. Some of these were very young, the smallest being about 1 mm. in size. In several localities in Nelson also I have found both young plants and prothalli, the habitat being in some cases a disturbed damp clay soil in the open, and in others a damp spot on the forest-floor. Old plants on several of these occasions were not to be seen in the vicinity of the young plants. Probably they had died out, or had been destroyed when the land was cleared. In not a few other parts of New Zealand I have seen young prothallial plants of this species both in southern-beech forest where the soil consists of decaying vegetable material and in shaded clay situations on open hillsdes. L. volubile would seem to propagate itself quite commonly by the dispersal and germination of its spores. The prothalli are bunied at a depth of ½–3 cm. Rarely they occurred at the surface, and then the upper part of the prothallus is a bright green. In an old gold-mining claim near Hokitika I found several prothalli both of this species and of L. scariosum. In both cases they were of an abnormally large size, and the young plants attached were as much as 6 in. in length and several times branched, and bore from one to three large adventitious roots.

L. ramulosum.

During the month of April, 1916, I found the prothalli and young plants of this species growing in an old shingle-pit on the side of the main road between Waimea and Kumara, Westland. In the near vicinity four species of Lycopodium occurred — viz., L. volubile, L. scariosum, L. fastigiatum, and L. ramulosum. This particular shingle-pit was in a damp but warm situation. There was an abundance of young plants of L. ramulosum of all stages of growth, and up to the present I have been able to dissect out from the mossy turves which I gathered twenty-four prothalli. The moss was a short brown variety, and when dissected was seen to be full of slime fungus and of several varieties of microscopic algae, with which the rhizoids of the prothalli were closely intermatted.

L. fastigiatum.

In one locality on a high exposed ridge near Mount Oxford, Canterbury, I have discovered sixty-one prothalli of this species during three years. Prothalli and young plants of both L. fastigiatum and L scariosum were growing on the leeward side of the highest point on this ridge in patches of moss and Helichrysum filicaule on clayey soil amongst rocks and boulders. The area in which the young plants occurred was very limited in extent, nor were they to be seen elsewhere along the ridge. The adult plants of these two species grew abundantly on Mount Oxford, which lies a mile or two to the windward of this ridge, but nowhere have I found them on the ridge itself. The great majority of the prothalli of L. fastigiatum were obtained from turves of clayey soil which came from the lower sides of embedded rocks and boulders.

– 262 –

In one instance I found a little group of fourteen prothalli actually touching one another in the humus underlying the moss, two of these prothalli being very small, not more than ½ mm. in size, and two others being old and exhausted. The prothalli were always buried to the depth of 1–5 cm., and occurred either in the humus underlying the covering of moss and Hechrysum or deeper still in the clayey substratum. I visited this spot frequently during three years, and on the last occasion found many very small and young prothalli. It was noticeable that the young plants invariably died after they had attained a length of 2 in. or 3 in. The patches in which the young plants and prothalli occurred did not die during the heat of the summer. All the ridges and hills in this neighbourhood were once covered with southern-beech forest and were still covered with the stumps of the trees, and it was ascertained that this particular ridge was burned off about fifteen or sixteen years before. Judging from the facts here stated, I am inclined to think that the spores were not blown from Mount Oxford, but had been shed from some mature plants which grew on the spot before the forest was removed. When the forest was burned the soil naturally would be very much disturbed and overturned, and favourable conditions would be set up for the germination of the spores and development of the prothalli. This would indicate that a long period of time is necessary for the development of sexually produced plants, a conclusion which is in accord with the observations of Bruchmann (6) in the case of the European species L clavatum, L complanatum, and L. annotinum. In several lowland localities near Hokitika, Westland, I have found young plants of L. fastigiatum.

L. scariosum.

I have collected about sixty prothalli of this species from two different localities on the Dun Mountains, Nelson, and from the Mount Oxford ridge mentioned above, and from a roadside clay cutting at Lake Kanieri, Westland, and have observed the young plants in various other localities. This species also would seem to propagate itself freely from spores when favourable conditions are present. The prothalli are always subterranean, and are more deeply buried than are those of L volubile or L fastigiatum, in some cases lying at a depth of 8–10 cm Several times. I have found the young plants of L volubile and L. scariosum growing together, and in one instance dug the prothalli also of both species from the same patch of soil. Both the prothalli and also the youngest plants of these two species are easily to be distinguished from each other, the former because they belong to different types, and the latter because in the case of L scariosum they are much stouter and coarser in appearance than in the other species, and also develop their charactestic form of heterophylly much earlier.

Summary.

The chief mode of propagation of the Lycopodiums is no doubt the vegetative one. Compared with other Pteridophytes, even with the Ferns, this characte is most extensively developed in the Lycopodiums. One finds large areas covered by some one species or other, as, for example, by L volubile, L. densum, L laterale, L fastgiatum, or L ramulosum, where in all probability the original individuals were the only ones which had been produced sexually. In this connection we notice that the form of most of the terrestrial species is well adapted to this end. It is long drawn out with branches of unlimited growth, in many cases the stems

– 263 –

and branches being subterranean, while at frequent intervals large adventitious roots are borne. It is in the epiphytic species that this character of extensive vegetative propagation is noticeably absent, and there, of course, the opportunity is wanting owing to the confined area in which the plants grow. The usual habitats of most of the terrestrial species are not suitable for the germination of the spores. Such forms as L. densum, L. fastigatum, L. volubile,* and L. scariosum luxuriate in open situations which are dry and unpromising. L. cernuum, L laterale, L. ramulosum, and L. Drummondi also grow most abundantly in peaty sour land which is alternately waterlogged or dried up, or even permanently wet. From my own observations, extended over a good number of years and in many parts of New Zealand, I have found it to be an almost invariable rule that young plants and prothalli are not to be met with in localities in which the adult plants are abundant. It is only in special localities, such as a damp shaded clay bank or roadside cutting, or some other patch of recently disturbed soil in the neighbourhood of adult plants, that the young plants occur. But it must be added that when the favourable conditions are present prothalli occur often in great abundance. In the case of several of the above-mentioned New Zealand species, prothalli and young plants were found in several localities in different years, while in certain species, as, for example, in L. cernuum, L fastigratum, L. volubile, and L. scariosum, they were discovered in large numbers. Bruchmann (6, p. 5) records that when once he had discovered the right kind of locality he was able to collect over five hundred prothalli of each of the species L. clavatum and L. annotinum. The epiphytic species grow under conditions which are normally more favourable to the development of the prothallus—namely, in shaded, damp, well-drained patches of humus. Treub (17) was able to find the prothallus of four different epiphytic species. Neither Miss Edgerley nor the present writer experienced great difficulty in discovering the prothallus of L. Billard. It is certain that in the case of some of the species—for example, L fastigiatum, L scariosum, and L. volubile—a long period of time, with consistently favourable conditions, is required before the prothalli are fully grown and the young plants are established. It has been shown above that there is reason to believe that a period of fifteen years may elapse before the spores of L. fastigiatum have germinated and the prothalli have developed. This is very similar to the conclusion reached by Bruchmann (6, p. 10) in those instances in which data to work upon were forthcoming. It would seem, however, that the prothalli of such species as L cernuum, L. ramulosum, and L. laterale, which are exceeding minute and delicate, require only a single season for their full development. These are the species which occur in a typically damp habitat. However, when the prothalli, and even also the young plants of terrestrial species, have succeeded in developing, a change of conditions, as, for example, a more than usually dry summer, may result in the destruction of all the young plants and of the prothalli also. Thus we conclude that although the spores of the Lycopodium species germinate freely under surtable conditions, yet the long period of the growth of their prothalli and young plants, and the uncertain conditions under which they live, have brought it about that the Lycopodiums have to depend mainly upon the vegetative mode of propagation. Probably we may also conclude that in the case of the terrestrial species propagation

[Footnote] * In parts of Westland, which possesses a wet climate, there are large breadths of this species where road cuttings have been made.

– 264 –

from the spore in now of importance only in the initial establishment of a certain species in new localities, as, for example, at the edge of the retreating forest, and that the subsequent peopling of these localities by such species, and their continued existence and spread, is due to vegetative propagation alone.*

III. External Features of the Prothalli.

L. Billardieri.

The prothallus of this species has recently been described in detail by Miss Edgerley (8, p. 104). External examination of my own specimens showed that it conforms more or less closely to the type of L. Phlegmaria. Miss Edgerley states that the internal structure also, except in one point, closely resembles that of the latter species. In each prothallus there is a central mass of tissue, in some cases compact and bulky, in others more extended and thin, from which long club-shaped processes arise (figs. 1–11, on p. 277). The prothallus is dingy white in colour, the ends

Picture icon

Figs. 1–6.Lycopodium Billardieri. Complete prothalli. 1, 2, 3, and 5, × 8; 4 and 6, × 4.

of the processes being more translucent in appearance Rhizoids arise over the whole of the prothallus except at the extreme ends of the processes Besides the hizoids, numerous short and thin processes, each of which is several cells in length but only one in thickness, arise from the surface of the prothallus. These will be probably the paraphyses described by Treub in L. Phlegmaria and by Miss Edgerley in this species. The youngest prothallus found by me (Fig 1) was 1 mm. in total length, and the largest from 12 mm. to 15 mm (figs 4–6).

[Footnote] *Where the land-surface was deeply covered by volcanic ash during the eruption of Tarawera in 1886, L. cernuum was growing in abundance some years ago near certain streams of hot water. Colomes of this description must have originated from spores, but, unfortunately, no exact data are available as to the first appearance of young plants.

– 265 –

L. laterale.

This prothallus (text figs. 13–16, and Plate XVII, Fig. 3) corresponds, with certain secondary differences, to the type of L. cernuum. All the prothalli found were colourless, except for a very faint tinge of green in the lobes. They are about 1 mm. in length. There is a lower rounded portion (“primary tubercle”) darker in appearance than the rest of the prothallus, and which in the solid stains more deeply with haematoxylin. Examination of serial sections showed that the cells of this region were occupied by a fungus. This lower part of the prothallus bears numerous long rhizoids. The middle region of the prothallus (“shaft”) in two cases was very short, giving the prothallus a stout and solid and opaque appearance (Fig. 14). In the other prothalli it was slightly longer, these appearing more cylindrical and drawn out. The leafy expansions on the crown of the prothallus are filamentous and less lobe-like than in L. cernuum. A characteristic feature of all the prothalli of L. laterale

Picture icon

Figs. 13–16.Lycopodium laterale. Complete prothalli, with young plants attached. 13, × 20; 14, × 10; 15 and 16, × 14.

examined was the presence of a filamentous outgrowth or a group of outgrowths on the shaft, either lower down towards the tubercle or higher up beneath the crown. None of the prothalli of L. cernuum seen by me showed any such leafy expansions on the shaft. Two prothalli, one of which is shown in Fig. 16, possessed a long process attached to the primary tubercle. The end of this process was club-shaped, and consisted of numerous cells, but the remainder was thinner, and showed two or three rows of longish cells. Goebel (10, p. 194) quotes from Treub that in L. salakense, and occasionally also in L. cernuum, several branches may arise from the primary tubercle. Perhaps the process found in the two prothalli of L. laterale is of the nature of such a branch. It is interesting, however, to compare this with the long-drawn-out

– 266 –

prothallus of L. ramulosum, described later. In the case of those prothalli which bore a young plant the foot could be plainly discerned through the semitransparent prothallial tissues. The prothallus of this species is very short-lived, and decays away after the young plant has developed two, or at the most three, protophylls.

L. cernuum.

Treub has fully described this prothallus. There are three regions—viz., the lower primary tubercle, the intermediate shaft, and the crown of lobes (figs. 17–21). In some cases the tubercle is pointed, in

Picture icon

Figs. 17–21.—Lycopodium cernuum. Complete prothali, with and without young plants attached. 17, 19, and 2, × 15; 18 and 24, × 18.
Fig. 21a.—Lycopodium cernuum Young plant, showing foot. × 18.

others rounded, at the lower extremity. I examined a large number of specimens, and noted that the chief variation in them was in the length of the shaft. This is dependent upon the depth of the germinating spore below the surface of the ground, for the crown of lobes is always at the surface Fig 17 shows one prothallus with an abnormally long shaft; on the other hand, in Fig. 20 it is seen that the shaft is almost absent. The crown in every case is very much lobed, the lobes of some prothalli being slightly greenish, and of others quite colourless. Rhizoids spring out of both the upper part of the tubercle and the lower

– 267 –

portion of the shaft. The tubercle is always thickly infested with a fungus, as also are the rhizoids. The necks of old archegonia were seen in surface view on the upper portion of the shaft immediately under the lobes. As in L. laterale, the foot of a young plant attached to a prothallus could be plainly seen through the tissues of the latter. The total length of the prothallus of this species varies between 1 mm. and 2 mm. No instances of branching from the tubercle were observed, such as has been described in the last species.

L. densum.

As already stated, no prothalli of this species were found, but judging from the fact that a large foot was seen on a young plant, which was over 4 in. in height, and that this foot was about an inch below the surface of the ground, it is concluded that the prothallus is large, firm, and long-lived, and more or less deeply buried.

L. volubile.

A description of the prothalli of this species and of L. scariosum has been given by Miss Edger (8, pp. 9599) from material supplied some years ago by the present writer to the Botanical Laboratory of the Auckland University College. The following account incorporates the chief points which I noted in the external examination of these prothalli and of others of the same species subsequently discovered. The smallest prothallus of L. volubile was about 1 mm in height (Fig. 22). There was a long tapering cylindrical projection below, and an upper more bulky

Picture icon

Figs. 22, 23.Lycopodium volubile. Young prothalli. 22 × 12; 23, × 15.
Figs. 24–27.Lycopodium volubile. Older prothalli. 24, × 6; 25, × 12; 26, × 8; 27, × 6.

portion. The prothallus was opaque and whitish in colour, except for its upper surface, which was more translucent in appearance. Fig. 23 is that of a slightly older specimen, in which a saucer-like depression suriounded by a thick rim is developing on the upper surface. In this prothallus the upper portion was above ground, and the translucent rim and depressed surface were green with chlorophyll. Still older prothalli are shown in figs. 24–28. In most of these it will be seen that the first-formed tapering portion is still present, in some cases as a blunt and in others as a pointed projection on the lower surface. The greater number of the prothalli showed a simple saucer-like depression on the

– 268 –

upper surface, surrounded by a folded rim, elongated in one direction. The rim was absent at the two ends of the depression, which, owing to the continued growth of the upper region of the prothallus, had in some cases become curved outwards and downwards. This agrees in all main points with the descriptions which Lang (14) and Bruchmann (6) have given of the prothallus of L. clavatum, and Bruchmann also in the case of the prothallus of L. annotinum. The oldest and largest specimens

Picture icon

Fig. 28.Lycopodium volubile. Prothalli in position relative to surface of ground. × 4.

sometimes showed one or more secondary folds on the upper surface, and rounded protuberances, which had the effect of almost filling up the original depression (Fig. 29). The prothallus shown in Fig 26 was growing at the surface of the ground, and its upper portion was coloured a dark green. Rhizoids are present on well-grown prothalli, especially on the region above the first-formed tapering portion immediately below the rim; but many of the prothalli seemed to be quite destitute of them,

Picture icon

Figs. 29, 30.Lycopodium volubile. A very large prothallus; top and side views respectively. × 2 ½.
Figs. 31, 32.Lycopodium volubile. Exhausted prothalli. × 8.

and were easily brushed clean of any adhering particles of soil or pieces of moss. In not a few cases two or three young plants were seen on the same prothallus. Figs. 31 and 32 show two exhausted prothalli, in which the long tapering first-formed erect-growing portion is very obvious, the upper part of the prothallus being flattish and slightly concave, and the rim sharp The depleted prothallus shown in Fig. 32, when viewed from below, showed between the tapering portion and the upper flat region a narrow ridge which corresponded in direction to

– 269 –

the elongation of the upper surface. This would seem to indicate that the horizontal extension of the upper region of the prothallus follows a more or less definite axis of growth. The largest prothallus found—that shown in figs. 29 and 30—was 6 mm. in length, the average size of a mature specimen being from 3 mm. to 4 mm.

L. ramulosum.

The figures illustrating the prothallus of this species are numbered from 32A to 32G inclusive. The youngest prothallus found was about 1 mm. in height, and is shown in Fig. 32A. It will be seen that in general form it corresponds with the young prothallus of L. inundatum figured from de Bary by Goebel (10, p. 191, Fig. 3). The lower rounded region was infested by a fungus, which extended somewhat up one side of the shaft. At its upper end the prothallus was extended into two

Picture icon

Fig 32A.Lycopodium ramulosum Young prothallus. × 20
Figs. 32B–32D.Lycopodium ramulosum Prothalli of long-drawn-out form, showing fungal regions, &c. × 12.
Figs 32E, 32F.Lycopodium ramulosum. Prothalli of stout, massive form, attached to young plants. × 12.
Fig 32GLycopodium ramulosum. The lower extremity of prothallus shown in Fig. 32F. × 35.
Fig 32HLycopodium ramulosum Protocormous rhizome, showing stem-axis, first root, and two vegetative bulbils. × 4.

processes, which had a slightly brownish appearance, as if withered. At the base of one of these processes was a darkly staining cell, possibly an antheridium. Below the upper end of the prothallus, on the under side of a short lateral process, three epidermal cells were seen to be infested by the fungus, and a rhizoid was borne on one of the adjacent cells. Rhizoids were also present on the lower end of the prothallus. Three young prothalli of this age were found. All the prothalli found were, in general, of the L. cernuum type, but they presented some

– 270 –

remarkable variations of it. In two or three instances old prothalli attached to young plants were very similar in appearance to those of L. cernuum shown in figs. 18 and 21. The other prothalli, however, were either long-drawn-out, being from 2 ½ mm to 4 mm in length, or short and comparatively massive. Eighteen prothalli of the former kind were found, two of which bore very young plants. Three of these are shown in figs. 32B, 32C, and 32D. Of the latter kind I have discovered six, all of which were attached to young plants. In figs. 32E and 32F are shown two of them. The prothallus shown in Fig. 32E possessed a lower rounded extremity, which bore hizoids, and whose cells were infested with fungus. It was evident, however, that this portion of the prothallus, which may be spoken of as the “primary tubercle,” since it corresponds to the region in the prothallus of L cernuum which has been given that name by Treub, is not the first-formed part of the prothallus, for its cells were continued below into an empty-celled process which had been broken off short. The shaft is very much longer than in any of the prothalli of L. cernuum or L laterale. Half-way up the shaft was a filamentous projection which bore sexual organs. On the opposite side of the shaft to this process two epidermal cells were infested by the fungus. At its upper extremity the prothallus was expanded into a bulky mass of tissue. The cells of the underneath region of this massive tissue showed the presence of fungus, and a group of rhizoids sprung from the same region. Immediately below the upper surface of the prothallus the necks of archegonia could be seen. Fig. 32C is that of a prothallus in which there were no fewer than five separate fungal regions along one side of the shaft. It was noticeable that each one of these regions was swollen and rounded, as if the presence of the fungus had served to stimulate the growth of the cell-tissue. A group of rhizoids was borne on each one of these fungal regions. Filamentous outgrowths with archegonia at their base occurred in two places on the ma shaft, opposite to the fungal zones. This prothallus was 4 mm in length. I did not observe whether it was growing upright or horizontal. All of the extended prothali were dissected out of short moss and slime fungus which was free from soil and decayed humus, and possibly the abnormal length of the shaft is to be put into connection with the depth at which the spores germinated below the surface of the moss, while the presence of several fungal regions beating hizoids is due to the absence of humus in the laye in which the prothalli were growing, and to the consequent dependence of the latter upon the intracellular fungus for much of the required food In Fig 32D is shown another prothallus, whose total length was about 4 ½ mm. It bore two groups of filamentous processes, at different places on the shaft. There were several fungal regions, all of which bore rhizords, and were slightly swollen, though to a less extent than in the prothallus shown in Fig. 32C. The four lower groups of fungus-containing cells were situated on the same side of the shaft, and were so close together that they almost formed one continuous zone. It will be noticed that in this particular prothallus the lowest fungal region is scarcely tubercular in form, and that it is continued below into a tapering empty-celled filament, which was probably the region of the prothallus first formed from the spore. The two uppermost fungal regions are on the side of the shaft opposite to the others. The long shaft is expanded above into a somewhat bulky crown, which bears filamentous and lobe-like foliar expansions. A diminutive young plant which had developed a single protophyll was attached to the crown of this prothallus, the

– 271 –

uppermost fungal region being situated immediately below it. In figs 32E and 32F are shown two solid massive prothalli. The latter of these was 2 ½ mm. in total height, and about 2 mm. in diameter in its upper region. These prothalli (and the other four of the same nature which were found) seem to correspond to the upper bulky region of the long-drawn-out prothallus of this species, described above. It would appear, however, that they owe their compact form not to the decaying-away of the shaft, but to its almost complete suppression. That this is so is evident from Fig. 32F, which shows the first-formed region of the prothallus in a remarkably intact condition. Fig. 32G is a much-enlarged view of the same. These short stout prothalli were in three instances found attached to young plants which bore four or five protophylls. Half-decayed prothalli were observed attached to young plants of still greater size and age. All the prothalli of L. ramulosum were green in their upper region and in the upper parts of the shaft. After studying the prothallus of this species I carefully examined again serial sections of several prothalli of L. cernuum, and in the case of one of them found that the primary tubercle was continued below into such a filamentous process as that shown at the base of the prothallus in Fig. 32B. I am inclined also to regard the club-shaped process described on the primary tubercle of two of the prothalli of L. laterale, and illustrated in Fig. 16, as corresponding to the first-formed region of the prothallus of L. ramulosum. It would seem thus that in these species, in some instances at least, and perhaps also as a rule, the spore on germinating gives rise to a delicate filament of cells which at some point or other soon becomes infected with the fungal element and then swells to form the so-called “primary tubercle” One prothallus found was shaped like the letter Y, it having branched into two more or less equal branches about half-way up the main shaft. At the point of branching the shaft was swollen and showed a fungal area, there being also the usual “primary tubercle” at the base of the shaft.

The prothallus of L. ramulosum presents some important features which serve to emphasize the great variability of the Lycopodium prothallus, and which suggest links between the different prothallial types much in the same way that the variations in form of the prothallus of L. Selago have been interpreted by Lang (14, pp. 3035). The form with the long shaft and scattered fungal areas indicates how the long-drawn-out prothallus of L. Selago, and so also the epiphytic type of prothallus, could have arisen from an ancestral L. cernuum-like type; and, on the other hand, the short massive prothallus of L. ramulosum, with its longer life and greater capability of the horizontal extension of its upper region, is intermediate in form between the L. cernuum type and the short variety of the L. Selago prothallus, and is suggestive of the subterranean L. clavatum and L. complanatum types.

L. fastigiatum.

The general form of this prothallus in the mature state would seem to be very like that of L. volubile, but there is an important developmental distinction to be noted. Figs. 33–40 illustrate specimens in which the manner of development of the adult form can be traced. The youngest prothalli found were about ½ mm in height. They showed a rounded tubercular opaque body surmounted by a smaller and more translucent region. Fig. 33 is that of a young prothallus 1 mm. in height showing these two regions. Very early in the development the

– 272 –
Picture icon

Fig. 33.Lycopodium fastigiatum. Very young prothallus, before branching × 15.
Figs. 34–38.Lycopodium fastigi [ unclear: ] atum. Young prothalli, showing stages in branching. 34 and 35, × 15; 36 and 37, × 12; 38, × 9.
Figs. 39, 40.Lycopodium fastigiatum. Older prothallus; side and underneath views respectively. × 10.
Fig. 41.Lycopodium fastigiatum. Full-grown prothallus, showing lobing on upper surface. × 5.
Figs. 42, 43.Lycopodium fastigiatum. Prothallus; underneath and side views respectively. × 7.
Fig. 44.Lycopodium fastigiatum. Large prothallus, with three young plants attached. × 5.

– 273 –

upper region becomes lobed (Fig. 34), and these lobes soon show themselves to be two main branches (figs- 35, 37, 38), each of which also develops secondary lobes: A large number of young prothalli of this age were collected, and they were all seen to show more or less clearly the Y form. In older specimens there is an extended and slightly concave flattened upper surface, the concavity being more cr less obscured by irregular lobing and folding (figs. 41 and 44). The original two main branches can always be distinguished in an underneath view of the prothallus (figs. 40 and 42). The view from above also in some case shows this, and shows as well, in spite of the complexity of the folding of the upper surfaces, that each main branch has divided into two branches, so that the upper surface is somewhat quadrangular in form (Fig 41). The first branching of the prothallus always takes place very early, so that in the older prothalli the peg-like projection is never as long and tapering in this species as it is in L. volubile. In the young prothalli the rhizoids arise especially from the upper ends of the two branches. The lobes are always clearer in appearance than the rest of the prothallus, which is in colour dingy white. None of the prothalli showed any trace of chlorophyll, being always more or less deeply buried.

L. scariosum.

Miss Edgerley (8, pp. 1002) states that the prothallus of this species resembles that of L. clavatum. I would unhesitatingly, however, say that it approaches more closely to the type of L. complanatum as described by Bruchmann (6). The material examined by Miss Edgerley had been collected by the present writer, and had been killed and fixed in 80 per cent. alcohol, and there is no doubt that it had shrunk considerably by the time that Miss Edgerley studied it. Hence she describes the upper surface of the prothallus as “concave …with a ridge running round the margin.” In figs. 45–53 in the present paper are depicted fresh prothalli of this species, and it will be see that they correspond very closely in general form to those of L. complanatum figured by Bruchmann on plate V and described by him on pp. 5759 of his work. There is, however, one main difference—namely, that old prothalli of the New Zealand species grow to a great size and become very irregular in form, so that a pseudo-branching is sometimes to be observed in them. The prothalli show a conical region below, in some cases blunt and rounded, in others more tapering, while above they are more massive. In the prothallus of this species there is no definite horizontal extension of the upper portion or lobing, such as there is in that of L. volubile. The upper region is surmounted by a bulky convex mass of tissue, which is semitranslucent in appearance, and is separated from the main body of the prothallus by a well-marked neck. In partially depleted prothalli, and in prothalli which have shrunk in alcohol, this upper tissue becomes slightly concave, but in fresh material it is seen to be invariably rounded and massive, and is not surrounded by a rim, but bulges out over the main body of the prothallus. In such an old specimen as figured in Fig. 53 this upper tissue constitutes the major bulk of the prothallus. A longitudinal section of a medium-sized prothallus, such as any of those shown in figs. 45–48, corresponds almost exactly with that figured by Bruchmann (6, pl. V, Fig. 25). My observations are, however, in agreement with Miss Edgerley's statement that the various layers of

– 274 –

fungus-infested tissue in the prothallus of L scariosum occupy a characteristically small proportion of the total bulk of the prothallus. Several old large prothalli showed a right and left portion separated by

Picture icon

Fig. 45.Lycopodium scariosum. Young prothallus. × 10.
Figs 46–48Lycopodium scariosum Prothall 46, × 6; 47, × 8.
Figs. 49–52Lycopodium scariosum Large irregularly grown prothalli. 49, × 8; 50, 51, and 52, × 5
Fig 53.Lycopodium scariosum Very large irregularly grown prothallus. × 6.

a cleft which extended through the upper translucent tissues well down into the body of the prothallus (figs. 51–53). One of Miss Edgerley's figures also illustrates the same feature This is probably to be regarded as a pseudo-branching resulting from the continued and regular growth in size of the prothallus, it being charactestic of the prothallus

– 275 –

of this species that it goes. on growing till it has quite lost its original form. The specimen shown in Fig. 47 was 8 ½ mm. in height. The old prothallus in Fig. 53 was 5 ½ mm. in height and 5 mm. in width, and no less than 13 mm (including the broken-off piece) in length. Fig. 49 illustrates an irregularly grown bulky prothallus, where the main lower portion has become divided into two by a horizontal constriction.

Summary.

In summing up this section I would lay stress first of all upon the fact that the study of the prothalli of the New Zealand species described above introduces no new type differing in any great degree from the five main types enumerated by such writers as Bruchmann, Lang, P [ unclear: ] tzel, Bower, and others. This is noteworthy, for it might have been expected that, since the discovery of the prothalli of only eleven out of the large number of modern species of Lycopodium had resulted in five distinct types being recognized, a considerable increase in our knowledge of the Lycopod gametophyte would reveal the existence of still further types. The prothallus of L ramulosum certainly presents remarkable features, but these are to be regarded probably only as variations from the L. cernuum type. Thus the results of the present study lend additional weight to the theory that in the sexual generation of the genus Lycopodium we are to trace the influence of two main factors—viz, epiphytism on the one hand, and a subterranean mode of life on the other—the development of the various types from a relatively primitive, chlorophyllous, and surface-living form having proceeded along these two main lines.

A second point to be noted is that although the New Zealand species of Lycopodium prothalli conform to the types so well known from the study of European and tropical species, yet in every case there are interesting modifications to be observed. This emphasizes the fact that the Lycopodium prothallus is a most variable one. But, further still, the nature of these various modifications may possibly be regarded as having significance in indicating how the epiphytic and the subterranean types have evolved from an original surface-living form, or even as supplying actual links connecting the different types together along two such lines of evolution. Lang and Bower both seem to consider the possibility of the genetic relationship of the different subterranean forms of prothallus, although they also suggest that they may be independent developments from a common ancestor. The most noteworthy variations from the ordinary types which occur in the New Zealand species of prothalli are as follows. The shaft of the prothallus of L. cernuum may be relatively long drawn out, suggesting the long cylindrical branches of the prothallus of L. Selago and of the epiphytic prothalli. The leafly expansions of the prothallus of L laterale a [ unclear: ] ie distinctly filamentous and more lanch-like than in the case of L. cernuum, and, moreover, these filamentous processes arise normally from other parts of the shaft as well as from its terminal region. The prothallus of L. ramulosum may be either long drawn out, with a somewhat massive crown, or short and bulky, with the shaft almost entirely suppressed and the crown greatly developed. Lang (14, p. 306) has also emphasized the variability of characters in the L. cernuum type which is known from the study of L inundatum and L salakense, such as the presence or absence of the crown of lobes and the manifold branching from the primary tubercle, features which seem to indicate a relationship between the L. cernuum

– 276 –

and the L. Selago types. It will be remembered that Thomas (16, p. 287) draws attention to the variation in the length of the shaft in the prothallus of Phylloglossum. Miss Edgerley (8, p. 107) in her description of the prothallus of L. Billardier has noted that it is simpler in organization than that of L. Phlegmara, in that the elongated central cells in the branches do not show the presence of pits on the walls, and that the paraphyses are only three cells in length and are never branched. The prothalli of L. fastigiatum and of L. volubile are in their mature form very much alike, being flattish and saucer-like; but, as has been described above, this form has been arrived at differently in the two cases. The branching of the prothallus of L. fastigiatum would seem to be reminiscent of the branching which is found in the L. Selago type, and also in the epiphytic and the L. cernuum types.

The close relationship between the form of the Lycopodium prothallus and its habitat, as illustrated in the species dealt with in this paper, must also be briefly noticed The prothalli of L. cernuum, L. ramulosum, and L. laterale are delicate, minute, and short-lived, and are surface-growing. The variation in length of the shaft of their prothalli is probably to be put in connection with the varying depth at which the spores germinate below the surface of the ground. These three species occui in wet habitats where the establishment of the young plant on the surface of the ground can readily take place. The epiphytic species of prothallus has an extensively branched and ramifying form Its typical habitat is a loose matted humus.. The greatest variety is to be found amongst those terrestrial species whose prothalli are subterianean. Some of these subterranean prothalli, such as those of L. volubile and L. fastiguiatum, occur chiefly in the loose humus in the top layer of soil immediately beneath the overlying moss and vegetable growth. These are the prothalli whose form is flat and saucer-like owing to the superposition of a horizontal mode of growth on the original vertical radial habit. The prothallus of L. scariosum, on the other hand, occurs as a rule much deeper and in more compact soil, and here the upright manner of growth is preserved and the saucer-like depression on the upper surface is absent. That view which would look upon the Lycopodium prothallus as a very plastic one would seem to be more in accordance with the foregoing observations than that other view which would regard the different types of prothallus as having been genetically distinct from a very remote period.

IV The Morphology of The Young Plant.

L. Billardieri.

The young plant takes its origin from the central mass of tissue of the prothallus. It first appears as a simple cylindrical stem (figs. 7 and 9), which it quite destitute of leaves for a length of 1–6 cm, according to the depth at which the prothallus is buried in the humus. The first-formed one or two leaves are generally scale-like, but the succeeding ones are laige and of the mature form (figs 10–12). A longitudinal section of the prothallus and young plant shown in Fig. 7 revealed the fact that the latter possessed a fairly large firm foot, and that there was a distinct epithelial layer of cells where the latter came in contact with the prothallial tissues (Fig. 8). It also showed the “first root” (throughout this paper this term is used to signify the first functional root) making its appearance as a protuberance at the base of the stem

– 277 –

The vascular strand of the stem and first root did not enter the foot. It was observed that the foot still persists on young plants which show as many as six to eight full-sized leaves. There was no trace of any swelling on the young plant which could be interpreted as the rudiment of a protocorm. The first root is developed relatively late. In the plants

Picture icon

Figs. 7–11.Lycopodium Billardieri. Complete prothalli, with young plants attached. 7, × 8; 9–11, × 4.
Fig. 12Lycopodium Billardieri. Young plant with foot and first roots. × 4.

shown in figs. 10 and 11 it had not appeared, although several full-sized leaves had been formed. All roots in the young plant subsequent to the first root are adventitious, and make their appearance on the stem above the foot (Fig. 12).

L. laterale.

In two previous papers (11, pp. 35761, and 12) the writer has given an account of the young plant of this species. The chief points in its development will here be given in reference to several microphotographs and drawings which are included in this paper. The prothallus of this species has already been described as belonging to the L. cernuum type. During the early stages in its develop-

– 278 –

ment the young plant is very similar to that of L. cernuum. It consists of a basal tuberous protocorm surmounted by one or two protophylls, and is connected with its parent prothallus by a foot which can clearly be distinguished through the prothallial tissues (figs. 14—16). The protocorm and protophylls are a vivid green in colour, and stomata occur on the latter. The protocorm bears numerous long rhizoids. From this stage onwards the young plant of L. laterale differs in its development from what normally takes place

Picture icon

Figs. 54–56.Lycopodium laterale. Young plants, showing beginning of extension of protocorm. × 16.
Fig. 57.Lycopodium laterale. Young plant shown in Fig. 56; side view. × 16.
Fig. 58.Lycopodium laterale. Young plant with four protophylls. × 16
Fig. 59.Lycopodium laterale. Young plant shown in Fig. 58; the latest-formed protophylls in end view. × 16.
Fig. 60.Lycopodium laterale Young plant shown in Fig. 58; underneath view. × 16.

in the case of L cernuum. The third protophyll arises in a lateral position (Fig. 54), and when it is full-grown its base shows as a swelling clearly to be distinguished from the original protocormous tuber (figs 55 and 56). In Fig 57 the young plant shown in Fig 56 is depicted in a sideways position, in which the two distinct swellings are clearly seen. The fourth protophyll arises alongside the third, and forms a pair with it (Fig. 58). Here again it is to be observed that the swollen bases of these two protophylls are distinct from one another. In Fig 59 are shown in end view the two latest-formed members of the plant illustrated in Fig 58 Fig 60 is an under view of the same plant, showing

– 279 –

the foot on the first-formed protocorm proper, and the bases of the two latest-formed protophylls. A transverse section of such a plant through the thickest region of the plant-body reveals clearly the original tuber and the later extension of it, but also shows that internally the swollen bases of the third and fourth protophylls are

Picture icon

Figs. 61, 62.Lycopodium laterale. Transverse section of two such young plants as shown in Fig. 58. 61, × 60; 62, × 35.

completely fused, although externally they can be distinguished (figs. 61 and 62). In no case was the prothallus found still attached to a young plant which bore more than the two original protophylls. The plant-body continues to grow sideways owing to the lateral development of new protophylls (figs. 63–65). In the majority of the young developing plants that were examined it was observed that the first-formed protocorm proper could clearly be distinguished from the later-formed protocormous extension, there being a well-marked constriction between

Picture icon

Figs. 63–65.Lycopodium laterale. Young plants, showing further extension of the protocorm. × 12.

the two parts (figs. 64 and 65). In the cleaning process the two portions in not a few cases broke away from each other. The protocormous rhizome continues to elongate laterally, owing to the further development of protophylls, till there are as many as eight to twelve of the latter. The protophylls arise in pairs, and in many cases they are arranged in two more or less distinct rows along the dorsal side of the rhizome. This, of course, indicates that their development, and so also that of the whole plant-body, has taken place very regularly. The rhizome is

– 280 –

covered on its ventral surface with a mat of rhizoids. In the earlier stages of its development it is bright green in colour and semi-translucent, but later it becomes yellowish and opaque and very firm. The

Picture icon

Fig. 66.Lycopodium laterale. Young plant with fully developed protocorm and a very young stem-axis. × 4.
Fig. 67.Lycopodium laterale. Young plant with a branched protocormous rhizome and two stem-axes. × 4.
Figs. 68, 69.Lycopodium laterale. Young plant with young stem-axis and first root just showing. 68, × 4; 69, × 5.

total length of the fully developed rhizome is from 3 mm. to 5 mm., and its thickness from 1 mm to 2 mm. In one instance it was observed to have forked into two equal branches, and on each of these a young stem-axis was developing (Fig. 67).

Picture icon

Fig. 70.Lycopodium laterale. Transverse section of fully grown protocormous rhizome before development of vascular strand, showing two protophylls above and groove on ventral side. × 30
Fig. 72.Lycopodium laterale. Portion of rhizome shown in Fig. 71, showing relation of vascular strand to the tissues of the protocormous rhizome. × 225.

The rhizome consists of parenchymatous tissues throughout, the cells of the central region being smaller and more compact, while those nearer the surface are larger and sometimes show air-spaces at their angles (Fig. 70) The centrally placed cells stain much darker with haema-

– 281 –

toxylin than those nearer the surface, owing to the presence in them of fairly abundant cell-contents. Each protophyll possesses a simple vascular strand, which penetrates into the upper region of the rhizome,

Picture icon

Fig. 71.Lycopodium laterale. Longitudinal section of fully grown protocormous rhizome, showing young stem-apex and course of vascular strand. × 25.

and there ends blindly (Fig. 70). A transverse section of a fully grown rhizome, such as is shown in Fig. 70, shows a well-marked groove running ventrally along the length of the rhizome. This is probably a consequence

Picture icon

Fig. 73.Lycopodium laterale. Longitudinal section of fully grown protocormous rhizome, showing initiation of first root. × 25.

of the fact that the body of the rhizome is formed by the swollen bases of the two rows of its protophylls. The stem-axis is initiated eventually at some point of the dorsal surface, and is marked by the

– 282 –

aggregation of several protophylls. This takes place either close to the growing end of the rhizome or at some point farther away from it, although the position of the young stem is always nearer to the growing end than to the first-formed protocorm proper (figs. 66–69, 71, 73, 75–78). Almost immediately vascular tissues are initiated from the stem-apex, and extend down into the upper region of the rhizome, receiving on the way strands from the neighbouring protophylls (Fig. 71). In the rhizome these vascular tissues bend round at an angle which is more or less sharp according to whether the stem-apex is farther away from the growing-point of the rhizome (Fig. 71) or close to it (Fig. 73), and take a course through the tissues of the rhizome, near to its dorsal surface, towards its growing end. At a later stage this vascular strand is surrounded by a slight zone of sclerenchyma (Plate XVII, Fig. 1). The arrangement of the

Picture icon

Fig. 74Lycopodium laterale. Portion of rhizome shown in Fig. 73, showing relation of vascular strand to the apex of root. × 225.

elements in the strand is described in the next section of this paper. Fig. 72 is a highly magnified drawing of a portion of Fig. 71, and shows that the vascular strand of the stem and first root as it passes through the body of the rhizome is developed from the actual tissues of the rhizome. The same fact is also apparent from Fig 74, which is a magnified drawing of a portion of Fig. 73, and from Plate XVII, figs. 1 and 2 At the same time that the vascular strand from the stem-apex is taking form along the body of the rhizome in the direction of its growing end, the latter at a point on the surface towards the dorsal side begins to grow outwards and downwards to form a finger-like protuberance (figs. 68, 69, 73, 74). Into this protuberance the vascular strand passes. This is the first root. The extension in length of the rhizome is brought to a close by its initiation. All subsequently formed roots in the young plant emerge adventitiously from the stem, and do not pass through the

Picture icon

Fig. 1.Lycopodium laterale. Transverse section of protocormous rhizome, showing relation of vascular strand to tissues of rhizome, × 40. (N.B.—In this figure and in the other plate figures which accompany this paper the cell have all been carefully outlined on the photograph with indian ink, in order to ensure clear reproduction.)
Fig. 2.—Lycopodium laterale. Longitudinal section of portion of protocormous rhizome, showing course of vascular tissue from stem-apex to first rot × 35
Fig. 3.—Lycopodium laterale. Complete prothallus. × 30

– 283 –

tissues of the protocormous rhizome. The leaves on the young stem-axis are in nowise different from the protophylls on the protocorm proper or on the rhizome; in fact, the ordinary vegetative leaves of the adult plant also have much the same form. The protocormous rhizome of this species is a persistent organ, owing to its large size and firmness. It may be recognized at the base of the stem of young plants which are 2 in. or even more in height.

L. cernuum.

The young plant of this species has been fully described by Treub, so that in this paper mention will be made only of such features as seem

Picture icon

Figs 75–78.Lycopodium laterale. Young plants with developing stem-axis and persisting protocorm. × 3.
Fig. 79.Lycopodium cernuum. Young plant, showing initiation of first root. × 12.Figs. 80, 81.Lycopodium cernuum. Young plants with developing stem-axis and adventitious roots. 80, × 6; 81, × 3.
Fig. 82.Lycopodium densum. Lower region of stem of young plant, showing foot and first roots. × 4.
Fig 83Lycopodium densum The foot shown in Fig. 82. × 40.
Fig 84.Lycopodium fastigiatum. Model of embryo of young plant, showing relative positions of foot and suspensor to the rudiments of the root, stem, and first leaf. × 105.

to be noteworthy for the purpose of a comparison with the young plant of the last species. The general form of the protocorm and first protophylls will be seen in figs. 17–19, 21, and 21A. In L. cernuum, as a rule,

– 284 –

the protophylls are all developed from the top of the protocorm, so that there is no noticeable lateral extension of the latter (figs. 79–81). In one or two instances, however, out of a great number of young plants examined I observed that there was a certain amount of lateral growth, the latest-formed protophylls occupying a position not on top of the protocorm, but at the side of it farthest from the foot. One young plant showed seven protophylls in all on such an extended protocorm before a stem-axis had been initiated. Plate XVIII, Fig. 1, is a photograph of a section of a young plant whose protocorm has thus grown sideways. In this particular plant the stem-axis had already been initiated on the end of the protocorm farthest from the first-formed portion, but it is not included in the section. The first root had also appeared, and is shown in the photograph. There is a distinct differentiation to be observed in the body of this protocorm between the protocorm proper and the laterally extended portion. As a rule, however, the protophylls are aggregated on the summit of the protocorm, and after a certain number have been developed—generally four or five—the stem-apex appears amongst them at the base of several rapidly growing new protophylls. At the same time that the stem-apex is initiated a finger-like protuberance appears in that region of the protocorm which lies immediately below the position of the stem, and grows outwards and downwards (Fig. 79). A longitudinal section of the young plant shown in Fig. 79 revealed the fact that vascular tissues were already developing from the stem-apex, and were declining into the root-like protuberance at a slight angle. A photograph of this is given in Plate XVIII, Fig. 2. The section shown in this photograph is not exactly median either for the stem-apex or for the root-apex, but it shows the course of the vascular strand, and also the position of the first root on the protocorm. The root-apex was protected by a cap of cells which had originated from periclinal divisions in the outer layer of cells of the apex. In the growing plant new leaves are rapidly developed from the stem-apex, and a stem-axis is formed (Fig. 80). Subsequently formed roots emerge either at the base of the stem through the tissues of the protocorm (Fig. 80) or higher up the stem (Fig. 81). The protocorm can be distinguished at the base of young stems which are from 1 cm to 3 cm. in height. As in L. laterale, the prothallus does not persist after the first two or three protophylls have developed.

L. densum.

As has been stated in a preceding section of this paper, I have not been able to discover more than a single young plant of this species. This was a plant about 4 ½ in. in height, erect, and branched above. One other developing plant was also discovered in the same patch of damp moss, but this was older, and had already assumed the creeping habit. These young plants occurred in the immediate vicinity of adult plants of L densum, and there were no other species of Lycopodium to be seen in the neighbourhood On the first-named “seedling” plant a large firm foot was present at a depth of about 1 in. below the surface of the ground (Fig. 82). The first root was borne at the base of the stem, and an adventitious root on the stem itself a short distance above the foot. Longitudinal sections through this foot failed to reveal any vascular tissue passing into it from the main stele. The main body of the foot consisted of large-sized parenchymatous cells, and there was

Picture icon

Fig. 1.Lycopodium cernuum. Longitudinal section of protocorm, showing course of vascular strand from stem-apex to first root. × 40.
Fig. 2.—Lycopodium cernuum. Longitudinal section of protocorm which had developed first root and stem-apex, showing unusually large size of protocorm. × 45.

– 285 –

a distinct epithelial layer of small regularly arranged cells (Fig. 83). There was also a central core of lignified cells which was continuous with the sclerenchyma zone of the stem. Although the prothallus of this species was not found, it is evident from the presence of the foot and its depth below the surface of the ground that the prothallus would belong to one of the subterranean types, as in the case of the three succeeding species, and that the young plant is dependent upon its parent prothallus for a considerable time.

L. ramulosum.

The young plant of this species corresponds very closely in its structure and manner of development to that of L. laterale. There is a protocorm which attains to a comparatively large size before the stem-axis is originated, and, as in the case of L. laterale, this protocorm may even branch. One peculiar feature, however, in the young plant of the present species must be referred to. In the young plant shown in Fig. 32H it will be seen that, contrary to what normally takes place, the growing end of the protocormous rhizome has continued its growth after the stem-axis and first root have developed. In fact, the end of this rhizome seems to have branched, the two developing branches appearing as bulbous outgrowths, each surmounted by a single protophyll. I am inclined to regard these bulbous branches as vegetative bulbils which would be capable of independent existence. In another instance seven young plants were found bunched together. They had to be extricated from one another. Two of them were exceedingly small, showing respectively one and two protophylls each; three others were of slightly larger size, having three or more protophylls; while the two remaining were much larger and apparently older, with somewhat irregularly grown rhizomes. Each of these plantlets or portions was developing, its growth being localized in some particular spot, which was a vivid green, while the lest of the plantlet was browner in colour. From the appearance of these plantlets and from the manner of their occurrence I concluded that they were vegetative outgrowths from, or portions of, one original and irregularly grown protocormous rhizome. In several other instances I have found two or more young plantlets in the closest proximity to an older and brown-coloured rhizome. These plantlets almost invariably consisted of a brownish basal portion surmounted by a few protophylls, one or two of which were semi-decayed, while at some point or other on the plantlet there was a bluntly rounded vividly green area which was obviously the growing region. In none of these instances was a prothallus or the remains of a prothallus to be seen anywhere near the young plants, although some of the latter were exceedingly small. Further investigation of this point is necessary, but it would appear to be probable that the protocormous rhizome of L. ramulosum under certain circumstances gives rise to vegetative bulbils which develop into young plants.

L. volubile, L. fastigiatum, L. scariosum.

In these species the prothallus is subterranean, and is large, firm, and long-lived, so that it supports the young plant till the latter has attained to a considerable size. In fact, two or even three plantlets may arise on the one prothallus in the case of each of these three species, although eventually only one continues its development. It will suffice to state that, as in the last species, and also in L. clavatum as described by Lang (14), the foot in the embryo plant of these three species is of a

– 286 –

large size, and consists of a uniform parenchymatous tissue bounded where it is in contact with the tissues of the prothallus by a distinct epithelium. In view of Miss Wigglesworth's statement (18) that in L. complanatum a short strand of vascular tissues passes from the main stele into the foot, I carefully examined transverse sections of the foot in several plantlets of different stages of growth of the present three species. In the case of L. volubile my results bear out very closely Miss Wigglesworth's statements. In the smaller plantlets of this species a strand of small thin-walled cells with abundant cell-contents penetrates well into the centre of the foot. It is in connection with the vascular tissues of the main stele. In older plantlets a few tracheides make their appearance in this strand. In one case, in the sections nearest to the main stele, these tracheides were in two groups separated by a single group of thin-walled cells, while towards the centre of the foot they gradually disappeared till there was only one left, much in the same manner as described by Miss Wigglesworth in the case of L complanatum In these larger plantlets the sclerenchyma in the foot closed in the vascular tissues together. In the young plants of L. scariosum and L. fastigiatum that I examined there is very little development of vascular tissues from the main stele into the foot. In the former species the foot is vary large, but only in the sections nearest to the main stele was any small-celled tissue seen to lead off from the stele into the foot, and no tracheides were present. In L. fastigiatum the foot-strand was slightly more developed, but not to the same extent as in L volubile. In all three species the epithelial cells of the foot remain intact even on the largest plantlets, but the outer walls of the cells become strongly thickened. The prothalli of these species belong to one or other of the large subterranean types, and they continue to grow in size in many cases long after the young plant has begun to develop. The development of vascular tissue in the foot of the young plant varies in extent in different individuals of the same species, and possibly this is dependent simply upon the size to which the parent prothalli grow There is no indication of a swelling comparable in any way with the protocorm of L. cernuum. The young plants take their origin from the upper region of the prothallus, in the case of L volubile and L. fastigiatum generally at one end or the other of the main groove or depression which is present upon the upper surface of the prothallus, and in the case of L. scariosum at some point or other on the margin of the upper bulging region. In each case the first root shows as a peg-like outgrowth at the base of the young stem The stems are cylindrical, colourless, and naked (except for a few minute scale-like leaves) for a greater or lesser length according to the depth of the prothallus below the surface of the ground. In L. scariosum, whose prothalli are the most deeply buried, the stems may be as much as 2–3 in in length before the ordinary assimilating leaves are formed. Young plants of each of these species are figured on the prothalli which have been referred to in Section III of this paper. The relative positions of the foot, stem, first leaf, and first-root rudiments in the young plant of L. fastigiatum are shown in Fig 84. This is a model of a developing embryo of this species viewed from above. It is tilted slightly in order to show the full sweep of the large foot and the projecting first root. In this case the apex of the stem and first leaf had just emerged from the tissues of the prothallus, but the root would not be apparent externally.

– 287 –

Summary.

The conclusions arrived at by Lang (14), Bower (4), and others with regard to the significance of the mode of dependence of the Lycopodium young plant upon its parent prothallus may here be briefly stated. Lang, following Treub, would regard not only the L. cernuum type of prothallus, but also the L. cernuum type of young plant, as being primitive for the genus. He says (14, p. 302), “The form of the young plant of L. cernuum, &c., is not to be regarded as recent and adaptive, but as possessing an important phylogenetic bearing.” With regard to the epiphytic species and L. Selago, he suggests that the protocorm stage has there been lost owing to the subterranean habit, and quotes Treub's statement of the existence of a rudimentary protocorm in L. Phlegmaria. In the subterranean types also of L. clavatum and L. complanatum the protocorm has been lost, and the large development of the foot is to be regarded as an adaptation in accordance with the large size of the prothallus and the lengthy dependence of the young plant upon it. Bower looks upon the L. Selago type of embryogeny, where the only extraordinary feature is the varying length of the hypocotyl, as being the most simple and primitive within the genus Lycopodium. In the types of L. clavatum and L. complanatum he would see an adaptation from the L. Selago type in accordance with the subterranean saprophytic specialization of the prothallus. He inclines to deny (3, footnote on p. 248) that the swelling in the embryo of L. Phlegmaria stated by Treub to be a rudimentary protocorm is to be interpreted as such. The protocorm of the L. cernuum cycle of affinity he would regard merely as a specialization, classing it lather as a parenchymatous swelling such as is the foot, and doubting any general application of the protocorm theory in the whole genus. The protophylls are, in accordance with his view, to be judged simply as turgid outgrowths from the protocorm. Goebel, in the first edition of his “Organography” (10, pp. 23133), does not favour Treub's theory of the protocorm, and lays emphasis on the fact that protocorms are found in certain epiphytic orchids. One other writer's views must also be briefly stated in order to bring forward the main lines of discussion along which the consideration of the Lycopodium embryogeny is being directed. Lady Isabel Browne (5, p. 223) has suggested that the protocorm is to be regarded as a modified form of stem due to reduction. She lays emphasis, on the one hand, on the great development of the stem in the oldest fossil Lycopods known to us, and, on the other hand, she concludes that “since vascular tissue penetrates for some distance in the protocorm of L laterale, this organ cannot, at least in that species, be dismissed as a mere parenchymatous swelling.”

No new facts of great importance emerge from the study of the morphology of the young plants of L Billardieri, or of L. volubile, L scariosum, L. densum, and L. fastigiatum. The foot of the young plant of L Billardieri seems to be larger than that figured by Treub for L Phlegmaria. In the young embryo plant of L. fastigiatum there is only one leaf-rudiment encircling the apex of the stem, and not a pair as figured by Bruchmann for L clavatum and L. annotinum. In L volubile. and to a less extent in L. scariosum and L. fastigiatum, a slight strand of vascular tissue is given off from the main stele into the foot.

But the study of the protocorm in L. laterale and L ramulosum brings to light facts of considerable interest. First of all, it must be emphasized that, as in the other species in which a protocorm has been found, it is

– 288 –

here associated with the L. cernuum type of prothallus. In these two species it would seem to be much more than a mere “temporary substitute for a root-system delayed in its development” (Bower), for it constitutes the plant-body for a whole season, attaining to a considerable size, and even branching. Moreover, the fact that vascular tissues develop within the main body of the protocormous rhizome suggests that in origin it is not a mere parenchymatous swelling. On these grounds it might be argued that the facts brought forward in this paper lend weight to the theory which regards the protocorm as a very ancient organ, possessing great phylogenetic importance. On the other hand, there are certain considerations which suggest that too much stress must not be laid upon the comparatively large size of the protocorm in L. laterale and L. ramulosum. In the first place, the manner of development of the protocormous rhizome in these species lends credence to the idea that it is merely a physiological specialization suited to carry the young plant over the dry season The original protocormous tuber, surmounted by its two protophylls, corresponds closely with that in L. cernuum, &c. The rhizomatous extension of the protocorm would seem to be, however, an added feature, to be interpreted apart from the original tuber. In its development it is markedly distinct from the latter, being separated from it by a constriction, and, moreover, it is initiated in L. laterale, and in certain cases in L. ramulosum, subsequently to the decaying-away of the prothallus, these two facts suggesting that a certain interval elapses before it begins to develop. The manner of growth of the rhizome strongly suggests that it is merely a specialized swelling, for it is the swollen bases of each new pair of protophylls which add to its length, and even the fully grown rhizome bears witness to the manner of its development in the arrangement of the protophylls in two more or less obvious rows on its dorsal surface and in the median groove running the length of the rhizome on the ventral side. Also, in L. cernuum a some-what similar lateral extension of the protocorm has been observed to take place, although to a less important extent, and this extension would seem to be a swelling distinct from the original tuber. There is a strong suggestion that the Lycopod protocorm is more plastic than an ancient and highly primitive organ would be expected to be, and that stress must not be laid from a phylogentic point of view upon the fact of its normally large development in L laterale and L ramulosum.

In the young plant of L. cernuum the vascular strand from the developing stem-apex takes a course through a corner of the protocorm. The short region of the latter which lies between the stem and the first-root apices may possibly be regarded as the rudiment of the stem-axis, retarded in its development and pushed out of its place through the intercalation of a tuberous stage. It is to be noted that the stem-apex and the first root always originate close together, and on the side of the protocorm farthest away from the prothallus. In L. Laterale, and in L ramulosum also, the stem and first root arise at the growing end of the protocorm, sometimes in close juxtaposition, though generally farther apart than in the case of L. cernuum. Here, too, it is possible to look upon that region of the protocorm through which the vascular strand passes as the rudiment of the stem-axis, very much postponed in development, and varying in size, owing to the intercalation in the embryogeny of the abnormally large rhizomatous swelling.

On the whole, the present writer inclines to the opinion that the large size and other abnormal features of the protocorm of L laterale

– 289 –

and L. ramulosum are to be regarded simply as a special adaptation, and would conclude that this lends weight to the theory that the Lycopod protocorm in general may best be interpreted in this way. The fact that the protocormous species are representative of three distinct sections of the subgenus Rhopalostachya, in which there is a considerable variation in the character both of the gametophyte and of the mature sporophyte generation, and is found also in the allied genus Phylloglossum, is significant, as indicating a considerable degree of antiquity for the protocorm within the genus Lycopodium.

V. Gross Anatomy of Stem and Branches.

L. Selago.

The vascular cylinder of the mature stem of this species in cross-section is stellate in appearance (Fig. 85). The xylem rows and protoxylem groups are stout, the latter being extended around the periphery of the cylinder in thick masses. At the base of the adult stem there are five to seven such protoxylem groups. The configuration of the vascular cylinder is variable: the centre may be occupied by xylem, or, as in the figure given, by an isolated group of phloem. The cortex is differentiated into three different zones; the innermost of these consists of cells whose corners are lignified, the outermost is developed as a broad zone of sclerenchymatous thick-walled tissue, and the middle cortex consists of a loose tissue of large thin-walled parenchymatous cells, through which several roots take their course. The vascular strand of the root generally consists of a large crescentic group of xylem embracing a single group of phloem; but in some roots there are two groups of xylem separated by a band of phloem; while in others, again, besides the main xylem groups there are a number of small isolated groups of protoxylem, in these cases the phloem being also in more or less isolated patches. A transverse section through the upper part of the stem shows the xylem in six stout rows radiating from a common centre, the protoxylem being in thick groups at their extremities. In this part of the stem there is no development of sclerenchymatous tissue in the cortex.

L. Billardieri.

In all the prothallial plants of this species shown in figs. 7—12 the vascular strand of the stem consisted of a single small crescentic group of protoxylem, enclosing between the horns of the crescent a single group of protophloem. In the plant shown in Fig. 11, which was about 1 ½ in in height, and possessed five leaves, the layer of cells immediately surrounding the pericycle had become slightly sclerenchymatous. The plant in Fig 12 showed two roots, each surrounded by a zone of well-developed sclerenchyma two cells in width, traversing the cortical tissues of the stem. The strand in these roots consisted of a single ciescentic group of protoxylem embracing a single group of protophloem. By the time the young plant has developed a dozen leaves the stem vascular cylinder shows two groups of protoxylem, which have been formed by the two horns of the crescent separating and the phloem extending between them. At this stage also the outermost layer of cells of the cortex has become slightly sclerenchymatous, as well as the layer immediately surrounding the pericycle. As the plant develops, the central cylinder becomes triarch by the splitting of one of the groups of protoxylem into two, the three groups of protoxylem alternating with three groups of phloem, the centre also being occupied by phloem.

– 290 –

Further development in complexity takes place by the splitting of the groups of protoxylem. In the young plants of this species, although the groups of protoxylem and the bands of phloem are compact and definite in form, yet from the very first a constant tendency towards rearrangement of the elements is apparent. The single crescentic group of protoxylem may split up into two or even three groups of single elements separated by single groups of phloem, and then join together again. And in the typical two-group stage the groups may be either compact and small or broad and extended around the periphery. The vascular cylinder in the mature stem (Fig 86) may be best described as stellate, and corresponds more or less closely with that described by Jones (13, p 23) for L. squarrosum, the chief difference being that in the New Zealand species the groups of protoxylem are much larger and the outer ends of the xylem rows stouter than in the other species. The phloem is in bands or in isolated islands, as also is the xylem. The configuration of the cylinder has a tendency to alter owing to cross-connections taking place, between the xylem bands and groups. The phloem consists of sieve tubes surrounded by a well-differentiated phloem parenchyma with abundant cell-contents. The bands of xylem vessels are not accompanied by any small-celled xylem parenchyma. The innermost and the outermost zones of the cortex consist of cells with more or less thickened walls, but it is only a layer or two of cells at both the extreme outer and inner edges which stain at all noticeably with safranin, and these only at the cell-corners. There is no marked rearrangement of the tissues of the vascular cylinder preparatory to a dichotomy. In one plant an exact trichotomy of the vascular cylinder was observed in the lower part of the stem. In the ultimate branchlets the number of protoxylem groups is reduced to four or three, the xylem being arranged sometimes radially and sometimes in isolated groups. In these branchlets the leaves, which throughout the plant in this species are comparatively large, are arranged in four to six orthostichies. In the fertile regions the sporophylls are always in four orthostichies, and the number of protoxylem groups is normally three.

L. varium.

In this species the stem is thick, but it consists almost entirely of thin-walled parenchymatous tissue. A narrow zone, five or six cells in width, immediately surrounding the vascular cylinder, is slightly sclerenchymatous. The central cylinder itself (Fig 87) is much smaller than in the last species. In the older parts of the stem there are seven or eight massive groups of protoxylem, and these are joined across by metaxylem or left isolated in a varying manner. The configuration of the stele thus cannot be definitely described. The groups of protoxylem not uncommonly join together as in the figure, and thus become greatly extended around the periphery of the cylinder. The cells in the centre of the phloem bands are empty, but are no larger than those of the phloem parenchyma which borders them on both sides. The latter have abundant cell-contents, as also do the cells in the pericyclic zone.

L. Drummondir.

The main stem here consists of a very loose parenchymatous cortex, the narrow innermost zone of which is sclerenchymatous, and a medium-sized vascular cylinder (Fig. 88). The metaxylem elements in the latter are arranged in not very compact bands, and sometimes single isolated

– 291 –

xylem vessels occur. The protoxylem is very much extended peripherally in narrow bands, several of which may join and so form a thin unbroken band extending a considerable distance around the cylinder. The phloem tissue is homogeneous, there being no differentiation into sieve tubes and phloem parenchyma. There is a pericyclic zone of cells lying between the vascular tissues and the sclerenchyma of the cortex, the innermost layer of this pericycle being composed of small phloem-like cells staining darkly with haematoxylin. The sporophylls are arranged in eight orthostichies in alternate whorls of four, but there is no correspondence whatever between the leaf-system and the configuration of the vascular cylinder in those parts. Both in the cone and in its pedicel

Picture icon

Figs. 85–96.Lycopodium, New Zealand species. Transverse sections of vascular cylinder of main stem. Semi-diagrammatie. Xylem and protoxylem elements indicated throughout. The circle in each case represents the inner limit of the cortex.
Fig. 85.Lycopodium Selago. Transverse section of vascular cylinder of lower region of stem. × 70.
Fig. 86.Lycopodium Billardieri. Transverse section of vascular cylinder of lower region of stem. × 40.
Fig. 87.Lycopodium varium. Transverse section of vascular cylinder of main stem. × 60.
Fig. 88.Lycopodium Drummondii. Transverse section of vascular cylinder of main rhizome. × 60.
Fig. 89.Lycopodium laterale. Transverse section of vascular cylinder of stout rhizome × 40.
Fig. 90.Lycopodium cernuum. Transverse section of vascular cylinder of main stem. — 20.

the characters of the vascular tissues of the main stem are reproduced, although on a smaller scale. It is to be noted that in this species a cone with its pedicel forms a not inconsiderable portion of the bulk of the whole plant, so that the vascular cylinder in the fertile regions is not greatly reduced in size.

L. laterale.

Fig. 89 depicts the vascular cylinder of a strongly growing main rhizome, such as is typical when this species occurs in an open situation. The inner cortex, comprising about one-half of the entire cortical tissues,

– 292 –

is slightly sclerenchymatous, but the outer cortex right up to the epidermis is thin-walled and spongy. In between the vascular tissues and the zone of sclerenchyma there is a pericyclic layer about three cells wide. The metaxylem elements are arranged in rows and groups, and the phloem in like manner. The latter is not differentiated into sieve tubes and phoem parenchyma, but is homogeneous. The protoxylem is in broad groups, very much extended around the periphery of the cylinder. In the sterile ultimate branchlets and in the cones the protoxylem is in three or four groups, which may be in connection at the centre or isolated: they are more or less extended peripherally. In the cones the sporophylls are arranged in six orthostichies in alternate whorls of three. There is no constant relation between the leaf-arrangement in those parts and the configuration of the vascular tissues. In a previous publication (11, p. 362) I stated that there was such a correspondence to be traced, but this statement must now be withdrawn, for I have observed that, while the arrangement of sporophylls does not vary, the vascular cylinder of the cone may be either quadrarch or triarch, according to the number of branchings that have taken place. A transverse section immediately behind the apex of the main rhizome shows a full-sized central cylinder, at the periphery of which there are tangentially extended protoxylem groups, while the centre of the cylinder consists of phloem and unthickened metaxylem elements. The differentiation of the metaxylem takes place in a regular manner from the protoxylem inwards. A transverse section of the protocormous rhizome of this species on which the young stem-axis and first root have appeared shows that the vascular strand as it passes along the protocorm consists of two broad groups of protoxylem separated by a single elongated group of protophloem. In the developing stem the vascular cylinder increases in complexity through the separation of the two original protoxylem groups into three and more, the phloem extending between them.

L. cernuum.

In the very young plant the vascular strand, which leads from the stem-apex through the protocorm into the first root, consists at first of a single small crescentic group of protoxylem which includes a small group of protoxylem between its two horns. Lower down in this first root the protoxylem separates into two groups and the group of protophloem extends between them, while higher up in the young stems the same takes place. A further stage in the young developing stem is reached by the two groups of protoxylem broadening out, and several metaxylem elements being formed between them, and at length joining the two groups together, thus dividing the phloem into two groups. The further growth in complexity of the vascular cylinder takes place by the protoxylem separating into three and more groups, the phloem extending between them. It is to be noted that from the very earliest stage the vascular elements do not preserve constant relative positions, but tend to separate easily from one another and then join together in a somewhat different arrangement. The vascular cylinder of the mature stem of this species has been described by Jones (13, p. 25), and the figure he gives corresponds very closely with that given in the present paper (Fig 90) Jones likens the vascular cylinder of L. cernuum to that of Gleichenia, and Boodle (1) describes that of the allied species L. salakense in the same way. The outer cortex of L. cernuum is sclerenchymatous, but the main inner bulk of the cortical tissues consists of thin-

– 293 –

walled parenchyma right up to the vascular cylinder. The metaxylem elements in the central cylinder are more irregularly disposed even than in L. laterale and L. Drummondii. Possibly it is more noticeable in the present species simply because of the much larger size of its cylinder. The protoxylem is extended peripherally, this being especially marked in the smaller stems and branches. In the main rows and groups of phloem the centrally placed elements are large, and have the appearance of sieve tubes, while those which surround them are smaller. There is no rearrangement of the vascular tissues preparatory to branching, but the main cylinder simply constricts into two more or less equal parts. As noted above in the last species, behind the apex of a main stem or large branch the differentiation of the metaxylem proceeds regularly from the protoxylem inwards; in these regions also the larger-sized phloem elements are seen to be empty, whilst the smaller phloem elements which surround them have abundant darkly staining contents. There is no definite radial arrangement of the vascular tissues in the ultimate branchlets or the cones.

L. densum.

The single “seedling” plant found, which was branched in its upper region, showed a radial arrangement in its vascular cylinder, there being six protoxylem groups connected in the centre by metaxylem and alternating around the periphery with six groups of phloem. Both the first root and the first adventitions root showed a triarch structure, the phloem in both cases extending between the groups of xylem and occupying the centre of the cylinder. The main rhizome of this species is stout and firm, and its vascular cylinder is among the largest in modern Lycopodiums (Fig. 91). The cortical tissues are throughout more or less sclerenchymatous, and increasingly so towards the centre. The middle region of the cortex is stored with starch. The xylem and phloem are arranged in alternate bands or plates, which lie parallel to one another in the plane of the ground. On the ventral side of the cylinder, however, this parallel arrangement is disturbed by the giving-off of adventitious roots. The division of the cylinder at a branching takes place right and left of a line at right angles to the plates of tissue. The number of groups of protoxylem is large, in the case figured there being seventeen, while at the base of some of the larger aerial branches the number may be as great as twenty-one. The xylem is differentiated into vessels and xylem parenchyma, and the phloem into large sieve tubes and phloem parenchyma. There is a pericylic zone of cells three or more in width. The aerial branches arise right and left of the rhizome, and then immediately turn upwards and grow erect. At their base the parallel structure is generally to be found, but this passes into the radial form higher up. In one strongly growing aerial stem, at a height of 4 ft. from the ground and below the first branching, it was observed that there were as many as twenty-one protoxylem groups, and that the configuration of the cylinder was markedly radial. Immediately behind the growing apex of the main rhizome the central cylinder is of the full size, and the differentiation of the metaxylem takes place from the protoxylem inwards. In the ultimate branchlets the number of orthostichies of leaves varies from six to eight, and the protoxylem is in three or four massive groups. A common condition is where the leaves are in six orthostichies in alternate whorls of three, and there are three groups of protoxylem. However, it cannot be said

– 294 –

that there is any constant relation between the leaf-trace system and the number of protoxylems, the latter in each ultimate branchlet depending upon the order of the branch. In the cones both the number of orthostichies of leaves and the number of protoxylem groups is variable the vascular cylinder may be either pentarch or quadraich. In the stouter regions of the adventitious roots there are from six to nine groups of xylem arranged round the periphery of the central cylinder, which are either joined up with each other or isolated. The protoxylem elements are very few in number. The innermost zone of the cortex is sclerenchymatous, as in the rhizome. The smallest roots show a single crescentic group of protoxylem, with a single group of protophloem. Larger roots show two and three groups of xylem, the phloem occupying the centre of the cylinder and extending between each of the xylem groups. The triarch condition is the common one in all medium-sized roots. It was noted that lateral rootlets are frequently borne in pairs, and that they arise by the trichotomous branching of the root-apex and vascular cylinder. A transverse section of a root at such a point shows that the vascular elements of each rootlet are derived from two adjacent groups of protoxylem and the intermediate group of phloem.

L. volubile.

The vascular strand in the stems of the youngest prothallial plants of this species examined always showed two small groups of protoxylem separated by a group of protophloem. The inner cells of the cortex from an early stage are slightly sclerenchymatous. The two groups of protoxylem in the developing stem join across, and the cylinder thereafter passes through a triarch and then a quadrarch radial stage through the splitting of the protoxylem groups. I have previously published an account of the development of the parallel arrangement of the plates of vascular tissue in the adult stem both of this species and of L. scariosum (11, pp 36264), in which the conclusion is reached that this arrangement is initiated by and persists from the branching of the stem in the plane of the ground. Also, the different mode of the development of heterophylly in the lateral branchlets of these two species has there been given. In L. volubile the first indications of heterophylly in a lateral branchlet do not make their appearance until the plant is 4 in. or 5 in in height (Fig 97). In slightly older plantlets all stages in the development of the heterophylly may be traced in the various branchlets on the same plant (Fig 98). In figs 99a and 99b is shown a portion of the dist [ unclear: ] chous region of a branch of a mature plant from the dorsal and the ventral sides respectively. The vascular cylinder of the main stem shows a more or less parallel arrangement of the xylem and phloem plates, this being disturbed, however, on the vent [ unclear: ] al side owing to the giving-off of the adventitious roots (Fig 92). There are from ten to sixteen protoxylem groups in the main stems. Each xylem plate is differentiated into vessels and adjacent xylem parenchyma, and each phloem plate into a row of very large sieve tubes bounded on either side by small-celled phloem parenchyma with abundant cell-contents. The whole of the cortex consists of thick-walled sclerenchyma. As in the other species which show the parallel disposition of the vascular tissues—viz. L. densum, L. scariosum, and L. fastigiatum—branching of the vascular cylinder always takes place in such a way that the division of the plates of tissue is in a line at right angles to the plane of their arrangement. In the large adventitious roots the configuration of the central cylinder

– 295 –
Picture icon

Fig. 91.Lycopodium densum. Transverse section of vascular cylinder of main rhizome. × 25.
Fig. 92.Lycopodium volubile. Transverse section of vascular cylinder of main stem. × 28.
Fig. 93.Lycopodium volubile. Transverse section of vascular cylinder of large adventitious root. × 40.
Fig. 94.Lycopodium ramulosum. Transverse section of vascular cylinder of scaly rhizome. × 40.
Fig. 95Lycopodium fastigiatum. Transverse section of vascular cylinder of main rhizome. × 30
Fig. 96.Lycopodium scariosum. Transverse section of vascular cylinder of main rhizome. × 25.

Picture icon

Fig. 97.Lycopodium volubile. Young erect plant, showing beginning of heterophylly. Natural size.
Fig. 98.Lycopodium volubile. Young plant with plagiotropic habit, showing different stages in development of heterophylly. Natural size.
Figs. 100, 101.Lycopodium scariosum. Young erect plants, showing early and sudden development of hetrophylly. Natural size.

– 296 –

is stellate (Fig 93), and there are about a dozen groups of protoxylem. As in the main stem, the whole of the cortex here also in sclerenchymatous. Branching of the adventitious roots and also of the lateral rootlets is by dichotomy, and not infrequently by a more or less exact trichotomy. As noted above for the last species, the smallest rootlets

Picture icon

Figs 99a, 99b.—Lycopodium volubile. Portion of distichous region of ultimate branch of mature plant, from dorsal and ventral sides respectively. × 4.
Figs. 102a, 102b.—Lycopodium scariosum Portion of distichous region of ultimate branch of mature plant, from dorsal and ventral sides respectively. × 4

show a single small crescentic group of protoxylem, while others show two or three xylem groups. In the ultimate branchlets there is no correspondence between the distichous habit and the number of protoxylem groups; the vascular cylinder in these regions of the plant is generally pentarch.

L. ramulosum.

In the stem of the young plant the xylem elements are more definitely coherent than in the case of L. cernuum, but the disposition of the xylem groups varies greatly. Both the cortex and the vascular cylinder of the more deeply buried scaleless stems of this species are larger than those of the scaly less deeply buried stems from which the aerial branches arise. In both the middle region of the cortex is sclerenchymatous, and in the more deeply growing stems the inner cortical zone is stored with starch. In Fig 94 is shown the vascular cylinder of the smaller scaly rhizome. In the larger rhizomes the number of protoxylem groups is generally eight or nine. The elements of the xylem are arranged in rows and groups, and do not show any tendency to separate from one another. There is no differentiation either of the phloem into sleve tubes and phloem parenchyma or of the xylem into vessels and xylem parenchyma. In the cones the sporophylls are arranged either in four orthostichies in alternate whorls of two or in six orthostichies in alternate whorls of three. There are four groups of protoxylem in the central cylinder of the cones, the groups being so broad and extended as to form the four sides of an almost complete square. The metaxylem elements, however, vary greatly in their arrangement.

– 297 –

L. fastigiatum.

In the stems of the youngest prothalhal plants examined there were two groups of protoxylem connected across by one or two elements of metaxylem, and a group of protophloem on either side of this median plate. In slightly older plantlets the number of groups of protoxylem is three or four, and they are joined together at the centre by the metaxylem elements. It frequently happens, however, that one or two of these protoxylem groups become isolated, and then two groups of the phloem become joined to form a band across the centre of the cylinder. In the full-grown rhizome the cortex is markedly differentiated into zones. There is an outer narrow zone of very slightly sclerenchymatous cells, which is separated from the inner main cortex by a narrow layer of very large thin-walled parenchymatous cells. The inner main cortical zone is strongly sclerenchymatous towards the centre. The vascular cylinder is large, there being from fifteen to eighteen groups of protoxylem (Fig 95). The xylem and phloem are disposed in parallel plates. This parallel arrangement, as in the other plagiotropic species which show it, is disturbed on the ventral side of the cylinder, probably in relation to the giving-off of adventitious roots. The xylem bands are differentiated into vessels and xylem parenchyma, and the phloem bands into sieve tubes and phloem parenchyma, the latter having abundant cell-contents. The pericyclic zone in this species is very broad, and is differentiated into an outer layer, three or four cells in width, whose cells are empty, and an inner layer, five or six cells in width, staining brown with safranin, whose cells show abundant contents. The arrangement of leaves on the cones is constantly in eight orthostichies in alternate whorls of four. The vascular cylinder of the cones also seems to have a very constant configuration, consisting of five groups of protoxylem joined together in a stellate manner, the centre of the cylinder being occupied commonly by an island of phloem.

L. scariosum.

In the young prothallial plants of this species the vascular cylinder is much stouter than in either of the last two species. The very smallest and youngest stems sectioned showed never less than three protoxylem groups, while both four and five groups are commonly found in the stems of young prothallial plants. The groups of phloem also are large, and the centre of the cylinder generally shows several large-sized elements of undifferentiated metaxylem. A series of transverse sections of a prothallus and young plant presented an opportunity for observing the changing configuration of the vascular strand as it passes from the stem into the first root in the neighbourhood of the foot. In this instance the young stem above showed five protoxylem groups. These groups in the region immediately above the foot joined up with one another around the periphery of the cylinder and separated again, regrouping themselves in a most irregular manner. One striking configuration was that of a somewhat irregular but continuous semicircle of protoxylem elements, on the periphery of the cylinder opposite to it there being a single compact group, and in the centre three isolated elements of protoxylem among the undifferentiated metaxylem elements. In the sections median through the foot the protoxylem groups were observed to be arranged in a triarch manner, but below it they adopted the form of a Greek omega, the root evidently preparing to fork. The characteristic heterophylly of this species [ unclear: ] makes its appearance very

– 298 –

early, even in plants which are still attached to their prothalli. In figs. 100 and 101 are shown two such young plants. It will be seen that, in marked contradistinction to what is found in the young plants of L. volubile, there are no gradual stages in the development of the heterophylly to be traced in these young plants, but the transition is sudden. In figs. 102a and 102b are shown the dorsal and ventral sides respectively of an ultimate branchlet in the mature plant of L. scariosum. The vascular cylinder of the mature rhizome is very large. The number of protoxylem groups varies from eighteen to twenty-seven, in the case figured (Fig. 96) there being twenty-three. The xylem and phloem are markedly disposed in parallel plates. The xylem bands show an absence of xylem parenchyma, and the xylem vessels are much smaller than in most of the other species described. In each phloem band the row of large sieve tubes is bounded on either side by phloem parenchyma, with abundant cell-contents. The entire cortex is more or less sclerenchymatous. In the ultimate branchlets, where the heterophylly is most marked, the configuration of the vascular cylinder is radial and quadrarch. In the cones the leaves are in eight orthostichies in alternate whorls of four, and the vascular cylinder is constantly hexarch.

Summary.

From the comparative study of the gross anatomy of the stem, roots, and branches of the eleven New Zealand species of Lycopodium the following leading facts stand out. In the most simple form, in the “seedling” stem and in the first root and all young rootlets, there is a single crescentic group of protoxylem embracing a single group of protophloem. In slightly older plants the vascular cylinder both of the stem and of the first adventitious root is either diarch or triarch, as it is also in most rootlets in the adult plant, and the number of protoxylem groups increases by splitting, so that a quardrarch or pentarch radial stage is attained. This is also the normal condition of the vascular cylinder in the ultimate branchlets and in the cones of the adult plant, except in L. cernuum and certain other allied species, in which the arrangement of the vascular elements is indefinite and mixed throughout the plant. In the ultimate branchlets the radial condition has resulted from the reduction in size of the mature vascular cylinder by sucessive branchings. The leaf-arrangement on the ultimate branchlets exercises no influence upon the configuration of the vascular cylinder in those parts, the number of protoxylem groups in the latter being dependent simply upon the order of the branch. In the main stems and branches and in the large adventitious roots the configuration of the central cylinder is stellate. The bands and groups of xylem and of phloem in the central region of the cylinder are more or less cross-connected, so that the actual centre may be occupied by a band or group either of phloem or of xylem, but at the periphery the disposition of both xylem and phloem is more or less ray-like. This form of vascular cylinder is found in species which belong to Pritzel's Selago and Phlegmaria sections of the genus. There are certain important modifications of this arrangement which are to be noted. In the first place, in certain species belonging to the sections Inundata and Cernua—viz., L. cernuum, L. laterale, and L. Drummondii—the protoxylem groups are very much extended peripherally, so that the vascular cylinder is more or less enclosed in a thin shell of protoxylem. This has also been described by Boodle (1) in the case of L. salakense. Moreover, the dis-

– 299 –

position of the xylem and phloem groups in these species is very mixed, and the individual elements cohere loosely together to form the rows and groups. This is most marked in the two species L. cernuum and L. salakense. It has been noted that in L. cernuum and L. laterale the vascular strand in the very young plant is indefinite in configuration, owing to the fact that it is largely made up of the leaf-traces from the neighbouring protophylls and the first-formed leaves. Moreover, in the adult stem of L. cernuum branching takes place not only right and left of the stem, but also from its dorsal side, while in the rhizome of L. laterale there is no regular plane of branching at all.

Another important variation from the typical stellate or radial vascular cylinder is the arrangement of the xylem and the phloem in more or less parallel plates. This is found in certain plagiotropic species which Pritzel has grouped in the Clavata and Cernua sections. It will be shown below that there are reasons for removing L. densum and L. volubile (both of which show the parallel character) from the vicinity of L. cernuum. Beside the two species just mentioned, L. scariosum and L. fastigiatum both show the parallel arrangement of their vascular tissues, and Jones (13) has enumerated various other plagiotropic species which do the same, all of which Pritzel has grouped in his section Clavata. This parallel character is due probably to the branching being restricted to the plane of the ground. In these plagiotropic species the growth in length of the main axis is unlimited and the growing apices are broad, so that the vascular cylinder at each branching is of the full size. Thus the parallel disposition of the vascular tissues, when once initiated, naturally persists.

The adventitious roots bear a marked resemblance to the rhizome in the plagiotropic species not only in vascular structure, but also in other characters. There is the same differentiation of the cortical tissues in both. The more deeply growing almost naked rhizome of L. ramulosum is interesting as suggesting an intermediate form between the typical scaly rhizome and the leafless adventitious root. The long aerial adventitious roots of L. volubile also might be termed either stems or roots.

The New Zealand species of Lycopodium vary markedly from one another in the differentiation both of their vascular and of their cortical tissues. The xylem elements may be either all of one kind, as also the phloem, or there may be a marked differentiation into vessels and accompanying parenchyma. Generally speaking, it is in the plagiotropic species that this differentiation is found in the greatest degree.

The view is here taken that the stellate or radially banded type of vascular structure, such as is found in the Selago and Phlegmaria sections of the genus, as also in the young plants and ultimate branches and the roots of most of the species, is the primitive one for the genus. The mixed L. cernuum type stands by itself, and has resulted perhaps partly from the dominance of the leaf-tree system over the cauline vascular tissues in the very young plant, and partly from the fact that along with the plagiotropic habit and the consequent fact of the presence of the vascular cylinder in its full size and complexity in the growing regions of the plant there is an irregular branching of the cylinder (branching taking place from all four sides of the growing apex), and hence a continual tendency to an indiscriminating mixing of its tissues. In the type which is characterized by the parallel arrangement of the vascular tissues we may recognize a modification of

– 300 –

the primitive stellate or radial form of structure, and one which is not so far removed from it as is the mixed type. Here, along with the plagiotropic habit, branching has become restricted to one plane, so that whereas in the young orthotropic stem the vascular tissues are arranged radially, in the mature stem a directive tendency is continually present at the growing apices towards a disposition of the bands of tissue in the plane of branching.

Jones (13, pp. 31, 32) would connect the epiphytic type of vascular cylinder with that of L. cernuum, but it seems to me from the study of the New Zealand species that the former is rather of the same nature as that of L. Selago. Pritzel's description in Engler and Prantl (9) of L. cernuum and L. laterale as erect-growing tree-like forms without widely creeping main axis is obviously incorrect, and his inclusion of L. densum and also of L. volubile in the section Cernua is open to objection. In her criticism of my former account of the anatomy of six New Zealand species of Lycopodium, Lady Isabel Browne has (5, pp. 22425) combated the suggestion that the mixed type of L. cernuum is to be regarded as relatively primitive. In this present paper it will be seen that I have accepted this criticism, and now speak of the L. cernuum type as a much modified one.

General Conclusions.

It remains now to bring together the main facts and conclusions noted in the different sections of this paper, and to attempt to estimate their value with respect to a natural classification of the genus Lycopodium, and also to put them into relation with the different theories which have been advanced regarding the interrelationships of the various species.

The prothalli of seven New Zealand species have been described—viz., L. Billardieri, L. cernuum, L. laterale, L. volubile, L. ramulosum, L. fastigiatum, and L. scariosum—while certain facts concerning the probable nature of that of another species—viz., L. densum—have been given. These prothalli have been found to belong to one or another of the types so well known from the descriptions of Treub, Bruchmann, and others that of L. Billardieri to the L. Phlegmaria type, those of L. laterale and L. ramulosum to the L. cernuum type; those of L. volubile and L. fastigiatum to the L. clavatum type, and that of L. scariosum to the L. complanatum type; while we may assume that the prothallus of L. densum will be found probably to belong to one or other of the two latter types. At the same time it has been found that interesting modifications are present in all of these New Zealand species of prothalli, such as may have significance in a biological or even in a genetic sense in supplying connecting-links between the various types. In L. volubile, L. densum, L. fastigiatum, and L. scariosum it was found that a large foot is a characteristic feature of the embryogeny, and in L. laterale and L. ramulosum an exceptionally large and long-lived protocorm has been described. The conclusion was reached from the study of its development that the large size of this latter organ in these two species is merely a physiological adaptation, and that this suggests that a similar interpretation is to be applied in the cases of the other species in which the protocorm is found. It was seen that a stellate of radial configuration of the vascular cylinder is characteristic in L. Selago, L. Billardieri, and L. varium, a mixed type in L. cernuum,

– 301 –

L. laterale, and L. Drummondii; and a parallel type in L. volubile, L. densum, L. fastigiatum, and L. scariosum. From the comparative study of the main stems, the roots, and the branches of all of these species, and of the development of the vascular strand in the young “seedling” plants of eight of them, it has been concluded that the radial or stellate type is probably to be regarded as most closely representing the primitive one for the genus, and that the mixed and the parallel types are modified forms of it.

In the main, the particular subdivision of the genus into sections adopted by Pritzel is in accordance with the facts known concerning both the sexual and the asexual generations of the various species. The Selago and Phlegmaria sections comprise species which are orthotropic when growing terrestrially, and pendulous when epiphytic. The prothalli of six of the species—viz., L. Selago, L. Hippuris, L. Phlegmaria, L. Billardieri, L. nummularifolium, and L. carinatum—are known: they belong to one or other of the two types, Selago and Phlegmaria, which both Bower and Lang would consider to be more or less closely related. The form of the embryo plant is simple, being uncomplicated by the presence of a large foot or a protocorm. The vascular cylinder in eleven species has been investigated by Jones (13), and also in L. Selago, L. Billardieri, and L. varium in the present paper, and has been seen to be constantly of the radial-stellate type. In the Inundata and Cernua sections occur those species whose prothalli are of the L. cernuum type, and whose young plants pass through the protocorm stage. This is known in five species—viz., L. inundatum, L. cernuum, L. laterale, L. ramulosum, and L. salakense. The structure of the vascular cylinder is mixed with extended protoxylem groups, this being especially characteristic of L. cernuum and L. salakense, and also of L. laterale and L. Drummondii; and Jones has noted the broad protoxylem groups in L. inundatum. Two species, L. volubile and L. densum, included by Pritzel in the Cernua section possess characters which possibly would justify their removal from this section. They both show the parallel structure of the vascular tissues, and both possess a large subterranean prothallus. The species in the Clavata section would seem to be closely connected with regard to their main characters. The prothallus has been found in five species—viz., L. annotinum, L. clavatum, L. fastigiatum, L. complanatum, and L. scariosum—and belongs to one or other of the two large subte [ unclear: ] rranean types, L. clavatum and L. complanatum. The large foot is characteristic of the embryo plant of these species. The vascular cylinder has been investigated by Jones in five species and by the present writer in L. fastigiatum and L. scariosum, and shows constantly the parallel structure.

The question arises, are these main sections of the genus which correspond so closely with the chief characters of both the gametophyte and the sporophyte generation to be regarded as more or less nearly related to one another, or as widely separated? The latter is the conclusion arrived at by Bruchmann (6, pp. 1089) from his study of the prothalh of four European species. Lang (14, pp. 3056), on the other hand, believes that the different prothallial types are “all more or less profound modifications of a type not unlike that of L. cernuum. The two forms of prothallus found in L. Selago give the clue” to the specialization of the subterranean saprophytic types on the one hand and of the epiphytic type on the other. With this view Bower (4) is in close agree-

– 302 –

ment. Various modifications of and variations from the main types have been enumerated by Lang, and others also, in this paper, which appear to supply connecting-links between them. The present writer would emphasize especially the variations in the prothallus of L. ramulosum in this connection. In respect of the character of the embryo types also, Bower and Lang both contemplate a certain degree of affinity between the different parts of the genus, with this difference in their views that the latter would look upon the L. cernuum type of embryo, with its protocorm, as primitive or the genus, the protocorm having been lost from the life-history of the species which belong to the other sections, whereas the former regards the L. Selago type of embryo as being the most primitive, there being derived from it the type which shows the large foot, while the L. cernuum type takes a middle position. In the present paper the conclusion has been arrived at from the study of the protocorm of L. laterale and L. ramulosum, and also of L. cernuum, that this organ, as Bower suggests, is a physiological specialization rather than a highly primitive feature. The comparative study of the vascular cylinder in the genus Lycopodium also suggests that the different sections as dealt with above are more or less closely related. Jones (13, pp. 2728) believes that a simple radial, root-like structure of the vascular cylinder is to be regarded as the primitive one for the genus, and my own view, derived from the study of the above eleven New Zealand species, is the same. The growth in size of this type of cylinder would either leave the configuration strictly radial or (on account of the constant changes in the disposition of the individual vascular elements and of the groups and rows in which they are arranged, a feature which is a characteristic one right through the genus) would cause it to adopt a stellate form in which cross-connections are present. Now, this is the type characteristic of the Selago and Phlegmaria sections. From this the mixed type of the Cernua and Inundata sections on the one hand, and the parallel type of the Clarata section on the other hand, would be derived, as has been described in Section V of this paper. Lastly, Miss Sykes (15) has shown that Pritzel's sections of the genus can be arranged in a series in respect of the morphology of the sporangium-bearing organs. From her study, however, she would conclude that the Cernua section comprises the most primitive members of the genus, and that those of the Selago section have been derived from them by reduction, while the type of L. inundatum is “an interesting intermediate step.”

Thus a belief in the interrelationship of the different sections of the genus Lycopodium is more generally held than the opposite one—that they have been widely separated from an ancient period. The view which the present study seems to suggest is that on a general sum of characters the Selago section must be held to comprise the most primitive members of the genus, and that the Phlegmaria and Clavata sections have been independently derived from it, the former being less modified than the latter. The exact relation of the Cernua and Inundata sections to the Selago section is difficult to gauge. The consensus of opinion seems to be that the L. cernuum type of prothallus is to be regarded as primitive for the genus, but the plagiotropic habit, the mixed type of vascular structure, and the protocorm, on the other hand, would all seem to be highly specialized characters. Hence the Cernua and Inundata sections may perhaps best be placed in a group by themselves, as having been derived from ancestors common to themselves and to the Selago section, but independently of the latter.

– 303 –

Literature Consulted.

1. Boodle, L. A. “On the Structure of the Stem in Two Species of Lycopodium.” Ann. of Bot., vol. xiv, pp. 315–17, 1900.

2. Bower, F. O. “Lycopods.” (Review of “Etudes sur les LycopodiacéAes,” par M. Treub, Part II, Annales du Jard. du Bot. de Buit., vol v.) Nature, vol. 34, pp. 145–146, 1886.

3. — “Imperfect Sporangia in certain Pteridophytes: Are they vestigial?” Ann. of Bot., vol. xv, pp. 225–67, 1901.

4. — “The Origin of a Land Flora.” 1908.

5. Browne, Lady Isabel. “Review: A Colonial Contribution to our Knowledge of the Genus Lycopodium.” “New Phytologist,” vol. xii, pp. 22–25, 1913.

6. Bruchmann, H. “Uber die Prothallien und die Keimpflanzen mehrerer europaischer Lycopodien.” Gotha, 1898.

7. Cheeseman, T. F. “Manual of the New Zealand Flora.” 1906.

8. Edgerley, K. V. “The Prothalli of Three New Zealand Lycopods.” Trans N Z Inst., vol. 47, pp. 94—111, 1915.

9. Engler and Prantl. “Pflanzenfamilien,” I Teil, iv Abteilung, Lycopodiaceae (Pritzel). 1900.

10. Goebel, K. “Organography,” Pt ii. 1st ed. Eng. Transl., 1905.

11. Holloway, J. E. “A Comparative Study of the Anatomy of Six New Zealand Species of Lycopodium.” Trans. N.Z. Inst., vol. 42, pp. 356–70, 1910.

12. — “Preliminary Note on the Protocorm of Lycopodium laterale.” Trans. N Z Inst., vol. 47, pp. 73–75, 1915.

13. Jones, C. E. “The Morphology and Anatomy of the Stem of the Genus Lycopodium.” Trans. Linn. Soc., series ii, Bot., vol. vii, pp. 15–35, 1905.

14. Lang, W. H. “The Prothallus of L. clavatum.” Ann. of Bot., vol. xiii, pp. 279–317, 1899.

15. Sykes, M. G. “Notes on the Morphology of the Sporangium-bearing Organs of the Lycopodiaceae.” “New Phytologist,” vol. vii, pp. 41–60, 1908.

16. Thomas, A. P. W. “Preliminary Account of the Prothallium of Phylloglossum.” Proc. Roy. Soc. London, vol. 69, pp. 285–91, 1901–2.

17. Treub, M. “Some Words on the Life-history of Lycopodium.” Ann. of Bot., vol. i, pp. 119–23, 1887.

18. Wigglesworth, G. “The Young Sporophytes of Lycopodium complanatum and L. clavatum.” Ann. of Bot., vol. xxi, pp. 211–34, 1907.