The Life History of the New Zealand Species of the Parasitic Genus Korthalsella.
[Read before the Otago Institute, May 9, 1933; received by Editor, May 29, 1933; issued separately, September, 1934.]
2. Material and Methods.
3. General Morphological Characters—
(a) K. Lindsayi.
(b) K. salicornioides.
(c) K. clavata.
4. Development of the Flowers—
(a) Female flowers.
(b) Male flowers.
5. Development of the Fruit.
6. Germination and Haustorial Activity.
Korthalsella is a genus of small parasitic plants including 15–20 known spp., which range from India and Malaya southwards and eastwards, to New Zealand and the Sandwich Islands. There are three New Zealand species—Korthalsella Lindsayi Engl., K. salicornioides Van Tiegh., and K. clavata Cheesem. The first mention of any of these plants comes in Hooker's “Flora Novae Zelandiae,” where Korthalsella salicornioides is described under the name Viscum salicornioides Hook. Hooker had seen only plants without flowers or fruits, and his description is of the vegetative parts only, though he includes a shrewd guess that the “flowers will probably be found to be very small and to be sunk in the tips of the joints: the perianth to be of four valvate petals with a cellular, porous, amorphous anther adnate to the face of each petal, the pollen lodged in the cells of the anther” (Hooker, 1853).
At this time the plant was known only from the Bay of Islands in the North Island. It was perhaps its rarity and evident peculiarity that led Hooker to indulge in such a detailed forecast. In the “Handbook of the New Zealand Flora” brought out by him eleven years later, K. salicornioides and Lindsayi are both listed as Viscum spp. (Hooker, 1864, p. 108). This time, with actual material to hand, Hooker was not nearly so detailed in his descriptions of the flowers. For Viscum Lindsayi he states: “Flowers very minute, whorled on the joints of the peduncle, perianth three lobed, lobes persistent,” and for Viscum salicornioides: “Flowers very minute, solitary or few together in the tips of the upper joints: perianth
three lobed, lobes persistent.” Korthalsella clavata was first discovered in 1876 by Kirk and Enys in the Castle Hill Basin, Canterbury, on Coprosma propinqua. The plant was neither in flower nor in fruit. In 1891, they found it again in the same district in another locality on Aristotelia fruticosa, this time in fruit. Kirk published a description of it in the Trans. N.Z. Inst., 1892. In the first edition of Cheeseman's “Flora of New Zealand” the three species are listed under the generic name Viscum (1906, p. 261): A note in the Appendix (p. 1151) refers to Van Tieghem's Memoirs on the Loranthaceae, in which are recognised many more genera for the family than are accepted in Engler and Prantl's “Pflanzenfamilien.” Unfortunately, I have not had access to Van Tieghem's Memoirs, and do not know the grounds on which he separated out the genus Korthalsella. In the second edition of Cheeseman's Manual (1925) Van Tieghem's generic name is adopted, though no reason is given for making this change, and we have Korthalsella salicornioides, K. Lindsayi, and K. clavata established.
Cheeseman follows Kirk and ranks K. clavata as a separate species, but includes the reservation that he personally suspects that it is merely a variety of K. Lindsayi. Miss L. M. Cranwell, M.A., botanist to the Auckland Museum, was kind enough to send the writer all Cheeseman's herbarium material and also some collected recently by herself at Castle Hill. This has been carefully examined, and the conclusion drawn that K. clavata is a distinct species intermediate in most respects between the other two.
The work on which this thesis is based was carried out in the Botany Laboratory of the University of Otago, Dunedin. The author wishes to express her thanks to Dr J. E. Holloway for his help and encouragement throughout.
Material and Methods.
Korthalsella Lindsayi and salicornioides are regarded as fairly rare plants. “Although they occur throughout the whole length of the North and South Islands of New Zealand, they are everywhere local and rarely occur in any great quantity.” (Cheeseman, 1910, p. 182.) Much of the hills around Dunedin, where the present investigation was carried out, is covered with Leptospermum ericoides scrub, but K. salicornioides occurs on the Leptospermum in only a few definite localities: it seldom occurs on anything but Leptospermum, though it has been reported also on Gaultheria and Dracophyllum. Similarly, K. Lindsayi, which occurs on a greater variety of shrubs—Cheeseman (1925) lists: Sophora, Melicope, Metrosideros, Myrtus, Coprosma, Myrsine, etc.—is found near Dunedin in the upper Botanical Gardens, on Helichrysum glomeratum and Melicope simplex, and also at one or two other localities on similar scrub. The infection in most of these places is dense, and there is no lack of material.
K. clavata is known only from Castle Hill Basin, where it is found on Aristotelia fruticosa, Discaria, and Coprosma.
Fresh material of K. Lindsayi and K. salicornioides was collected, and killed and fixed on the spot. Inflorescences for the various stages of floral and fruit development were fixed in Carnoy's Fluid, using 95% instead of absolute alcohol. Chromacetic was used also for a few collections, but was found to be not so good as the Carnoy. Seedlings, stems, and haustoria were fixed in formalin alcohol. The material was imbedded in paraffin and cut at 6, 8, and 10 μ. Double staining with safranin-gentian violet, and safranin-light green, and triple staining with safranin-gentian violet and orange G, were all brilliant after these killing and fixing reagents. Some of the later stages of fruits, with mucilaginous tissue, were difficult to infiltrate. The hard twiggy host stems with haustoria were found not to cut successfully after imbedding in paraffin. These were soaked in glycerin for two to three weeks to soften, and then cut in pith with a sliding microtome. This procedure gave very satisfactory results. Hand-cut sections were used for various micro-chemical tests.
Flowering material was collected from the end of August onwards. By the beginning of November it was found that the young fruits were forming. Seeds and seedlings were collected from January to April.
All drawings other than those showing general habit were made with the aid of either an Abbé camera lucida or a Zeiss reflex drawing apparatus. In the case of the habit drawings dividers were used for accurate measurements.
General Morphological Characters.
Korthalsella Lindsayi: Figs. 1 and 4 show the habit of the plant. It is small, branched, succulent, and perfectly glabrous, jointed at the node, which is restricted, while the internode is flattened and broadly obovate. The outer part of the internode is continued up into a collar round the node. The growing apex is protected by these collars, which fold over it in just the same way as would young leaves, but each collar is a ring of tissue, not a whorl of separate emergences. At the node inside the collar, branches and inflorescences arise, usually 2 from one node, but sometimes 3 or 4. There is a thick cuticle, and most of the epidermal cells are strongly lenticular on the outer walls, especially in the young parts. Only the very youngest parts of the plant look green. The older branches and even the inflorescences look brown. Sections of the tissue show that in the photosynthetic layer, which lies under the epidermis of all parts of the plant, are groups of cells filled with a yellowish brown substance. Hand-cut sections of fresh material were treated with FeCl3, and also some with FeSo4, to see if these cells contained tannins. There was no colouration. With osmic acid this substance did not blacken. It is probably then resinous. Some sections were soaked in alkannin tincture for three days, but there was no detectable reaction either: this may have been due to the rather unsatisfactory nature of this stain, or to the resinous cells being already too coloured to show any change. Fig. 13 is a diagram of a transverse section of an internode, showing the position of these groups of resin cells.
As observed at Dunedin, K. Lindsayi infects Helichrysum glomeratum and Melicope simplex very densely. For example, on one branch of Helichrysum glomeratum there were 184 separate K. Lindsayi plants and also 5 plants on the K. Lindsayi itself. The obovate internodes of the Korthalsella are much the same shape as the small leaves of the two host plants mentioned, and this fact, combined with its more or less brown colour, makes it very inconspicuous unless carefully looked for.
The flowers are borne on definite inflorescences which are terminal, or arise as lateral branches from the nodes (Fig. 5). The flowers, which are green or brownish, are arranged in a definite way. At each node of an inflorescence there are 2 groups of flowers separated by 2 tufts of hair. Each group consists of one male flower, which stands out well above the collar, and four female flowers in a row below it, with only their tips projecting above the collar (Figs. 9 and 10). These groups are arranged regularly one above the other at successive nodes, up and down the inflorescence. Fig. 11 represents a longitudinal section cut in the plane of the rows of male flowers. In some instances a decussate arrangement has been noticed: at one node the groups will alternate, as in Fig. 9 at the tip. This variation occurs fairly frequently. In one inflorescence it was found that at one node both the groups were doubled, consisting of 2 male flowers and 6 or possibly 8 female flowers per group. This was quite exceptional.
The vascular system consists of two separate main collateral bundles in the internodes, more or less fusing in the constricted node (Fig. 13). The separate bundles traversing the internodes give off nervures which supply the photosynthetic layer just under the epidermis. All the elements in the bundle have very small lumina. The tracheids are short, and are intermixed with the xylem parenchyma. This structure explains the brittle nature of the plant, which is very easily broken across. Vessels are not so common as very short tracheids, and are very narrow and mostly quite short also. The writer has not been able to distinguish any typical sieve tubes, and doubts if there are any. It seems that their place is taken by parenchyma—elongated cells, with rich cytoplasmic contents and conspicuous nuclei. In the older parts, at the base of the plant, the bundles are united to form a loose vascular cylinder—Fig. 36. Fig. 35 represents a single bundle in transverse section.
Stone cells occur at any point in the plant and are to be seen in many of the figures. Crystal-containing cells occur commonly in the outer part of the flowers. These crystals are insoluble in dilute acetic, but dissolve in dilute nitric acid without the evolution of gas bubbles, and so were shown to be probably Calcium oxalate. These are common features of the Loranthaceae. (Solereder, 1908.)
K. salicornioides is a small tufted plant with slender terete branches, branched to a greater extent than K. Lindsayi. Fig. 2 is a habit drawing of K. salicornioides parasitic on Leptospermum ericoides. The stems appear jointed, being slightly constricted at each node. The outer photosynthetic and succulent part of the stem
is continued as a collar, which ensheathes the actual node and the lower end of the next succeeding internode. This structure is exactly analogous with that in Salicornia, where it has been suggested that the outer succulent part of the stem really represents the modified leaves from the node below, the collar then representing the fused and modified tips of these leaves. The ensheathing collars, just behind the apex, where the internodes have not yet elongated, fold over, and protect the growing point just as in K. Lindsayi: this has the same effect as a series of imbricating leaf primordia.
The vascular system is almost identical with that in K. Lindsayi. The internodes are traversed by two main bundles which give off strands into the photosynthetic tissue. Fig. 14 is a diagram of a transverse section of an internode. At the node, the bundles give off branch strands to the flowers or fruits and the lateral branches. In the oldest regions of the stems at the base of a plant there is, as in K. Lindsayi, a loose vascular cylinder. It is noticeable that all the vascular elements are narrow.
Infection of the Leptospermum by this species is sometimes very dense, and in some cases a host branch is clearly affected detrimentally by the infection. Such a branch may be seen to bear dead-looking, leafless twigs. In one instance 88 distinct plants of the parasite were counted on a single branch of Leptospermum.
Stone cells and crystals of Calcium oxalate occur throughout the plant as in K. Lindsayi. The epidermal cells also protrude strongly and have a thick cuticle. Cells completely filled with a bright orange-brown, supposedly resinous, substance occur dotted in the photosynthetic layer (Fig. 14). The same micro-chemical tests were applied to these as were carried out for K. Lindsayi with the same results.
Unlike K. Lindsayi there are no definite inflorescences, the flowers being borne at the nodes towards the tips of the ordinary branches. Here, also, they occur in the same definite group formation of one male flower and four female together, there being two groups to each node. The arrangement of the groups at the successive nodes is alternate and opposite—a decussate arrangement—in contrast to the regular vertical rows in K. Lindsayi. Fig. 8 is a habit drawing of the tip of a branch bearing flowers, and Fig. 12 a diagrammatic drawing of a longitudinal section of the same. Whereas there were irregularities in the arrangement of floral groups observed in K. Lindsayi—an alternate arrangement occurring in places—in K. salicornioides no irregularities were found. As a large number of different inflorescences have been examined, it seems safe to conclude that probably the salicornioides arrangement is the more normal, and represents the type from which the K. Lindsayi arrangement has arisen.
In the Manual of the New Zealand Flora (Cheeseman, 1925) the floral descriptions for the species of Korthalsella are very imperfect. This no doubt is due to the fact that those descriptions are based on material representing a late stage of development in which fruit formation was well advanced. At the time the female flowers are fully developed only their perianth tips are above the
collar. They are, indeed, exceedingly minute for fully developed flowers, but sections show the embryo sacs to be ready for fertilisation. The male flowers are considerably larger and stand out well above the collar over the top of the female flowers (Figs. 8—12). Cheeseman (1925) states that the male flowers are much smaller than the female flowers, indicating that fruit formation was well advanced in the particular shoots he examined. He failed to notice the regular arrangement of the flowers: as the fruits grow larger and project above the collar, the arrangement becomes obscured, especially as all the flowers may not develop, or some may be injured or knocked off at some intermediate stage, so that gradually the definite numbers and the regular arrangement is lost. In K. salicornioides the male flowers persist for a surprisingly long time; they are usually to be found squeezed in between the fruits, which are clustered round the node. In K. Lindsayi they drop off early after shedding their pollen.
Korthalsella clavata is shown in Fig. 7. It has the general appearance of K. Lindsayi, but theinternodes are more attenuated. The most definite distinction is that the flowers are borne in whorls at the upper nodes and not on definite inflorescences as in K. Lindsayi, though the upper internodes between the topmost whorls of flowers become quite small and terete, as they are in the K. Lindsayi. A comparison of Fig. 5 and Fig. 7 brings out this distinction. This feature, as well as the more slender nature of its vegetative parts, determines K. clavata to be quite distinct from K. Lindsayi, and worthy of a distinction of a separate species. On the whole, it is intermediate in character between the other two New Zealand species. The habit is like K. Lindsayi, but more slender, while the flowers are borne as they are in K. salicornioides. All the material that the author has seen was in a well-advanced fruiting stage, and no flowers could be found. Thus it was impossible to determine whether or not the arrangement of floral groups is decussate or in vertical rows.
Development of the Flowers.
The flowers, which are morphologically practically identical in the two species, K. Lindsayi and K. salicornioides, grow out as a whorl of undifferentiated ovoid cell masses in the axils of the collars behind the growing tip.
The female flower (Fig. 15a) is quite globular, three or, quite rarely, four pegs of perianth lobes appearing at its tip as the flower nears maturity with the sessile stigma between them. One or more embryo sacs appear in the centre of the solid mass of the flower. Fig. 20 is a drawing of a median longitudinal section of a mature female flower of K. salicornioides. The embryo sac is not cut medianly. It is seen that the flower is extremely small. It is quite sessile. There is no differentiation whatever into the typical floral organs. The sessile stigma is merely a receptive spot on the tip of the flower; its cells are secretory and the stigma is sticky, pollen grains becoming stuck to it. The perianth pegs are the minute protuberances from the tip of the flower encircling the stigma.
Vascular strands, consisting of one row of trachea elements, run up the side of the flower out into each lobe. In the mature female flower there is a suggestion of different cell layers. The cells outside the vascular strands are fairly larger. Those round the embryo sacs are fairly compact, except for those which directly abut on them, and which are becoming used up as the embryo sac grows. In later stages of fruit formation there is a more marked differentiation into layers. In the phylogenetic development of the flower each of these layers has probably been derived from a different floral organ. The position of the embryo sac with reference to the perianth lobes is inferior. The outer cell layer of the flower probably represents the fused bases of the petals, the three perianth lobes representing the free tips of these, not the complete reduced petals. Inferior ovaries occur in many different families of plants, but all are not morphologically identical. Johnson (1889) and Thoday and Johnson (1932) have discussed reduction of the female flower for two species of Arceuthobium, Treub (1882) for two species of Loranthus and one of Viscum, and Van Tieghem (1869) for Viscum album. From these it is clear that the Viscum female flower is more reduced than that of Loranthus. Treub brought forward evidence in favour of the derivation of the former type from the latter. The floral type in Korthalsella seems to the present writer to be even more reduced than that of Viscum.
In Korthalsella one or more spore mother cells develop in the central dome-shaped mass of compact tissue, and a tetrad division takes place in each. Linear tetrads can be seen, in which either the uppermost or the lowermost cell is developing into an embryo sac. Fig. 23 shows a linear tetrad. Embryo sac formation appears to be quite normal. Several embryo sacs may reach an advanced state in each flower. The number found varied from one to five. There is no sign of any definite arrangement of these, as has been noted in Viscum album (Treub, 1882). They may lie side by side in a group or separated by parenchyma. At exactly what stage a single embryo sac becomes dominant is hard to say. In the youngest stages of fertilised embryo sacs only one is left, the others having degenerated. The cytoplasm in the mature embryo sac is richly stocked with highly refractive globules. Although the flower is so reduced that the organisation of even the ovule is lost, there is, so far as these investigations show, no reduction in the embryo sac.
The male flower is perhaps the most peculiar feature of the plant. Figs. 15b, 8, 9, and 10 show the outward appearance with three perianth lobes—much bigger than the mere pegs on the female flower—folded over a rounded head in the centre of which is a pore, through which pollen (poll.) is being shed; the pollen grains adhere in a mass. Cheeseman (1925) states that the male flower has sessile anthers adhering to the perianth lobes. This is quite incorrect; he probably thought that the flower was the same as in Viscum album, but this is not the case. Fig. 30 shows a median longitudinal section of an unopened male flower of K. Lindsayi. Fig. 27 is a drawing of a slightly oblique transverse section of an open male
flower of K. salicornioides, and Fig. 29 shows a longitudinal section of an open male flower of K. salicornioides. The complete male flower has only one pollen sac, which is divided by septa into six pollen chambers. This whole structure is annular in shape and arises from the base of a short perianth tube. The central opening is a very minute pore, so that the outward appearance of this annular structure is hemispherical. The pore leads down to the central part of the flower or receptacle, which is separate from the sac, though it abuts closely on it. The wall of the pollen sac consists of an outer epidermis, a subepidermal fibrous layer, and one inner layer of cells: this wall is continuous from the circular outside margin at the base of the petals, over the top, down the neck of the central pore, and forms the bottom wall of the sac; the fibrous layer is not differentiated in the bottom wall, but seems to extend for a longer or shorter distance down the neck (Figs. 27 and 30). There is a separate tapetum round each of the six pollen masses as Fig. 28, which represents a tangential longitudinal section, clearly shows. In an open flower both the tapetum and the inner layer of cells have been used up, and also the epidermis outside the fibrous layer has disappeared, so that the wall of the pollen sac at this stage consists solely of the fibrous layer. The septa break and the pollen chambers become continuous. The circular wall round the pore below the end of the fibrous layer become ruptured, and an opening is thus made from the pollen chambers through the pore to the outside. The pollen exudes in this manner from the pore (Figs. 9 and 10). No stages were observed in the present study younger than that shown in Fig. 30. Possibly the manner in which the ring-shaped pollen sac actually originates would afford an indication as to how it is to be interpreted. In the light of what is found in the other members of the family, it would seem possible that it represents fused sessile anthers adnate to the fused petal bases. In Fig. 27 a bending in of the fibrous layer at the septa between the pollen chambers in two places may indicate the boundaries of fused sessile anther lobes.
The condition of the male flower in other Korthalsella spp. should help towards an interpretation of this structure in K. Lindsayi and K. salicornioides. I have not had access to Van Tieghem's Memoirs on the Loranthaceae and do not know if an account of any of these other species is given.
Development of the Fruit.
The fertilised egg is scarcely distinguishable from the endosperm cells which completely fill the embryo sac from a very early stage (Fig. 21). The embryo is an undifferentiated cell mass which grows very slowly at first, all cells resulting from divisions of the fertilised ovum going to form the embryo which soon becomes separated from the sac membrane by a layer of endosperm cells (Fig. 26). The endosperm divides fairly rapidly and enlarges at the expense of the surrounding cells. At the same time the cells of the fruit wall show a differentiation into layers. In the middle layer, from the
tip of the fruit to the depth of the endosperm, but not beneath it, the cells become first meristemmatic (Fig. 21), then elongate radially (Fig. 26). As the embryo and endosperm grow, these cells become differentiated as the mucilaginous elements, the individual cells becoming greatly enlongated. Their nuclei also become attenuated and stain with sanfranin as red streaks. The cell contents disappear, and the walls finally become mucilaginous, the whole mass staining very densely (Fig. 31). The embryo lies at the top (anterior) end of the endosperm and when the fruit is ripe protrudes from it (Fig. 32). The embryo cells are small, with large nuclei and dense cytoplasm, the endosperm cells densely stocked with starch granules. The embryo is surmounted by a cap of parenchymatous tissue. When the seed is shed, it consists of endosperm, embryo, and the few cells of the cap, the whole surmounted by the viscid tissue. It is an unprotected structure, lacking a true seed coat. It is very small and quite unadapted to bird attraction.
Many fruits fall off complete, the embryo, endosperm, and adherent mucilaginous cells becoming more or less shot out of the fruit coat through the broken end of the latter. In some fruits the tip is ruptured and the seed forced out through the top, no doubt by the swelling of the mucilaginous tissue. A certain internal pressure is indicated by the flattening of the innermost cells of the fruit wall (w in Fig. 32), but the fruit is so small that this is not very great. In all cases the fruit or the seed falls near at hand, many adhering to the parent plant itself. This highly inefficient seed distribution renders the plant incapable of spreading itself over any distance. The seeds are produced in large numbers in proportion to the size of the plant, and these are brought to maturity between November and February. Each contains a considerable amount of stored food in the form of starch granules (Fig. 32), more than it would seem possible for the plant, with its restricted photosynthetic apparatus, to manufacture in the time. It is highly probable that the source of this supply is the host, from which carbohydrates as well as water and salts are taken. The path of this translocation is most likely through the xylem, where there is a large connection between haustorium and host, and not through the phloem, in which it is very doubtful if there is any connection with the parasite.
Germination and Haustorial Activity.
The embryo is already protruding slightly from the endosperm when the seed is shed (Fig. 32). It immediately grows out to form a radicle which closely applies itself to the surface of the host. The radicle seems to digest its way through the outer layers of the host and grows into the vascular tissue, with which it at once forms connections. As the young seedling develops, a growing point becomes differentiated under a cotyledonary collar, which remains for a time embedded in the endosperm from which it absorbs the stored food (Fig. 19). In K. Lindsayi the seed usually sends out a short radicle which immediately applies itself to the substratum.
In K. salicornioides the radicles observed were longer and seemed to straggle along the stem until they grew into a niche in the bark (Figs. 17 and 18.) Many seeds fall on to leaves of the host plant, and when these are shed the seeds are lost. Commonly in K. Lindsayi they become lodged in the angles of the host stems. Seedlings of this species were found to occur much more plentifully than those of K. salicornioides: the former were also found not infrequently on the parent plant (Figs. 3 and 4), though this occurrence is as much a result of the inefficient seed distribution as of the dense growth. Fig. 18 shows a K. salicornioides seedling on the parent plant. When the K. Lindsayi seeds become lodged in the angles of the branches of the parent plant they germinate readily. The young seedlings are readily distinguishable from the substratum by their light green colour, small size, and attenuated hypocotyl and radicle region, but at a somewhat later stage they are no longer distinguishable from an ordinary branch. No doubt if a careful investigation of many plants were made and sections cut of the branching it would be found that many K. Lindsayi plants are compound, consisting of two or more, parasitic the one upon the other. When, on the other hand, the seed becomes attached to a flat internode its character as a plant parasitic on its parent does not become obscured with age (Figs. 3 and 4). In K. salicornioides, the seeds were found to be not so commonly in the angles of the Leptospermum stems as along the twigs, where they stuck readily to the rough bark. The number of seedlings of this species found later was not in proportion to the number of seeds that had stuck and begun their germination. The rough bark is continually flaking off, so that, besides those seeds which drop off or are knocked off, a number will be shed with the bark before the radicle of the seedling has grown down into the host stele. Notwithstanding, on some branches of Leptospermum the K. salicornioides was seen to be exceedingly plentiful.
The young haustorium penetrates the host and spreads out in a club-shaped head. Fig. 37 shows a longitudinal section of a K. Lindsayi stem just above a node with a young haustorium penetrating; the haustorial head is indicated by stippled cell contents. The head of the haustorium enlarges by digesting the tissue immediately in front of it. Very soon bands of haustorial tissue begin to grow along the line of the cambium, both up and down the stem, and round the side of the stele. Fig. 36 is a transverse section of a K. Lindsayi stem just above a node showing a seedling haustorium. The haustorium has sent arms round the cambial line of the loose vascular cylinder. Tracheidal cells in the haustorium directly abut on the tracheidal elements of the stem stele. The cells of the haustorium along its margin are large, with large brilliantly-staining nuclei. These cells appear to be actively digesting the tissues in front of them. The extent of the haustorium in Fig. 36 is indicated by showing the cell nuclei. Fig. 38 is a drawing of a very young haustorium of a K. Lindsayi seedling on a K. Lindsayi internode, the latter being cut transversely. Haustorial cells all were thin-walled and had large nuclei, much larger than
those in the large-celled part of the host stem, though not a great deal larger than those of the photosynthetic layer. The haustorium is tapping the elements (C.V. and S.V.), which are crushed by its growth and have been separated from their fibre (f) by a considerable amount of secondary growth of the haustorium (2°). The haustorium entered the bundle from the side at the cambium layer. It digested the cambium and soft tissues of this bundle, and has since enlarged considerably in its place. A comparison of this figure with Fig. 35 of a single bundle will make this point clearer. The isolated vascular elements, S.V., probably belong also to the xylem of the original bundle and have been surrounded by haustorial tissue. Fig. 33 shows a transverse section of a Leptospermum twig with an older haustorium of K. salicornioides. The haustorial tissues are drawn in detail; the tissues of the host twig are indicated by cross hatchings. The section is taken where the Leptospermum twig is giving off a branch trace, so that there is a certain irregularity as seen, for example, in the contour of the pith. The haustorium penetrated along the line of the cambium, which, as well as the phloem, it has completely eliminated. No phloem was to be seen at any point on this section between the outer margin of the haustorium and the periderm. The haustorial cells became meristemmatic, and a considerable amount of mixed parenchyma and short tracheidal cells has been formed. These are all linked up with the vascular system in the Korthalsella stem. At the side of the host stem into which the haustorium first grew there is the greatest amount of haustorium developed. Towards the far side of the stele there is an increasing amount of wood. The cambium here was the last to be eliminated, and would continue to be formed until the haustorium grew right round. The difference in the amount of wood at the near and far sides, however, represents a longer time than is represented in the age of the parasite plant, so that a fair amount of digestion of actual woody elements has taken place. Again the margin of the haustorium consists of large papillose cells which appear to be both forcing themselves between the elements of the host and to be digesting them.
Fig. 34 shows in detail a portion of a band of haustorial cells in a very young haustorium of K. salicornioides on a Leptospermum twig. The cambium has been digested away, and also the very young wood and the young phloem. Tannin in a wood medullary ray and in a few inner bark cells is black. The thick line along the phloem margin of the haustorial cells was a deeply staining strip of material, doubtless indicating active digestion along that border. Cellulose tissues are drawn black; lignified tissues are cross-hatched. The nuclei in the haustorium were large and brilliantly staining. The ovoid bodies in the haustorial cells are probably chloroplasts. In one series of sections of a medium, aged haustorium on a Leptospermum twig the vessels of the host were seen to be all occluded with resins or tannins along the advancing margin of the haustorium. The cells, which were being killed by the secretion of the parasite, were forming resistant masses in the tissues. Perhaps the mistletoe
killed the wood parenchyma and the wood rays faster than it could digest them. Their material broke down into these substances just as it happened in the case of cells killed by exposure in wounds or cells cast off in the bark.
Korthalsella salicornioides and K. Lindsayi are greatly reduced. This reduction is exhibited in the habit generally and in the very small size of all the parts; in the structure of the flowers in which there are none of the typical angiosperm floral organs; in the lack of leaves; in the general simplicity of organisation of the plant parts, and also in the completely undifferentiated embryo. The photosynthetic tissue is of small extent in proportion to the number of fruits matured, and is largely blocked by resins in the older stems. This must mean that the Korthalsella draws manufactured food from the host—that parasitism has proceeded to a further degree than in those leafy Loranths which need to draw only water and salts.
The great reduction in size of the fruit has led to restricted and localised occurrence of the plants, because all the seeds are shed very close to the parent plant, and many on to the parent plant itself. On K. Lindsayi with its flattened internodes, this “cannibalism” occurs to a much greater extent than in K. salicornioides. This complete inability in both species on the part of the plant to distribute itself over any distance has resulted in its becoming confined to isolated districts. Where it does occur, however, it may be exceedingly plentiful. An indication of the slowness with which distribution takes place is seen in the frequent fact that some individual shrubs can be densely covered with one or other of the parasites, whilst others of the same kind within a few feet are quite uninfected.
This type of reduction to extreme minuteness of the whole plant and of its fruits has become definitely disadvantageous. It seems that these species of Korthalsella have passed the minimum of reduction and must be on the road to extinction. They are no longer able to distribute their seeds any distance or actively to invade new ground. Their distribution throughout the length of New Zealand, but everywhere rather rare and local, also points to their being dying species.
Korthalsella is not being eliminated by any competition. It is dying out of its own inefficiency. There is no competition between the host and the parasite, though there may be a certain amount of competition between several plants parasitic on the same branch. This competition cannot be a very strong factor in the survival of the Korthalsella when infection can continue up to the stage it does. Another factor that may influence its survival would come into operation if the extent of the infection prevented the further growth of the branch or absorbed so much of its food supply that it died. This probably does occur, but rarely, the writer never having seen an instance of a branch killed by Korthalsella on it, although many infected plants have been examined. This fact is of greater significance in
larger parasitic species which are more of a drain on their hosts. Competition between infected and uninfected plants does not occur; if uninfected plants are near enough to compete they become infected by contact with the plants they would be likely to oust. Where the Leptospermum scrub becomes invaded by other shrub species, as sometimes happens, a new factor in the elimination of the Korthalsella by the elimination of its host by non-susceptible species comes in.
These reduced characters of Korthalsella salicornioides and K. Lindsayi can scarcely be interpreted as adaptations to anything. It is sometimes considered that parasitism induces reduction. In most cases parasites are reduced or peculiar plants. It is argued that with a ready food supply the stimulus to form food-manufacturing organs, namely, leaves, is removed. But it is not easy to see how a plentiful food supply should influence the formation of flowers, so that gradually the organisation of these should become altered.
In the Loranthaceae there are many peculiar and highly specialised species which represent, as it were, the twigs of a branching family tree; the relationship between some is close; between others more distant. Several lines of descent can be traced, especially in floral morphology, yet in all these species the general habit is the same, and the type of environment similar also. In all families different lines of descent can be traced. Where the changes of structure cannot be explained in terms of adaptations to an environment it is said that there is a trend of development in a family, that it is a case of orthogenesis. There are many characters in plants, especially floral characters, which are totally inexplicable as adaptations and on which natural selection could not have been conceived to act, for example, the structure of the male or female flower in Korthal-sella. Buchholz (1) has pointed out a case of selection which is often overlooked, and which is of wide occurrence in Angiosperms as well as in Gymnosperms. In Korthalsclla there are several embryo sacs to each flower, but only one embryo to each fruit. A domination of one individual has occurred, perhaps we might say through a metabolic character which enabled it to grow quickest. Every individual has a whole complex of characters, many of which are directly linked. At most of the critical stages in the life history there is only one, or at most only a few, controlling factors. It is at the critical stages of the life history that selection occurs. Characters which thus influence but one stage, e.g., rate of metabolism controlling growth of fertilised embryo sacs will be preserved, together with any characters to which they have become linked. It is thus conceivable how useless and even detrimental characters may become preserved through linkage to other characters. The evolution of the plant must occur at all stages of the life cycle. The survival and continuance of useless and detrimental characters in an orthogenetic trend in any line of descent, as the reduction of the flower in the Loranthaceae, is possibly to be explained by the continuance of a whole string of inter-related characters, of which enough are useful in the life history of the plant
to carry the others which are not useful or even disadvantageous. The balance between useful and detrimental characters is the margin on which the species progresses or retrogresses.
1. The New Zealand species of Korthalsella were first put in the genus Viscum by Hooker. Cheeseman in his 2nd Edition of the New Zealand Flora moved them into Van Tieghem's genus Korthalsella.
2. There are three closely allied species. They are small tufted parasitic plants 2–4 inches high, leafless and glabrous, with jointed stems, terete in K. salicornioides, and with flattened internodes in K. Lindsayi and K. clavata. The flowers are minute, borne in groups of one male and four female flowers together, two groups forming a whorl at a node.
3. The female flowers are ovoid with three minute perianth pegs at the tip, a sessile stigma between. They consist of homogeneous tissue with 1–5 embryo sacs in a group at the centre. The male flower, also minute, has a single six-chambered pollen sac which is ring-shaped, and is considered to represent fused sessile anthers adnate to a short perianth tube. It is a highly peculiar structure.
4. All cells resulting from divisions of the fertilised egg form the embryo, which is an undifferentiated globular mass of cells; a viscous layer of radially elongated mucilaginous cells develops from the middle region of the fruit wall. When the fruit is ready to be shed the embryo has elongated and is protruding from the endosperm.
5. Dehiscence is variable, the seed slipping from the fruit to fall near at hand; seed distribution is ineffective.
6. Germination frequently takes place when the seed is shed on to the parent plant, resulting in a sort of cannibalism.
7. The genus Korthalsella is considered to be very reduced. The typical floral organs are not represented, the flowers being very simple and very minute. The small size of the fruits has resulted in inefficient seed distribution leading to localised occurrence of the plants. It seems that the New Zealand species of Korthalsella have passed the minimum of reduction and are on the road to extinction.
Buchholz, J. T., 1922. Developmental Selection in Vascular Plants. Bot. Gazette. 73, p. 249.
Cheeseman, T. F., 1906. Manual of the New Zealand Flora, 1st Ed.
—, 1925. Manual of the New Zealand Flora, 2nd Ed.
—, 1910. Contributions to knowledge of the New Zealand Flora, Trans. N.Z. Inst., 43, p. 182.
Hooker, J. D., 1853. Flora Novae Zelandiae.
—, 1864. Handbook of the New Zealand Flora.
Johnson, T., 1889. Arceuthobium Oxycedri. Ann. Bot., 2, p. 137.
Kirk, T., 1891. On a New Mistletoe, Trans. N.Z. Inst., 24, p. 429.
Solereder, H., 1908. Systematic Anatomy of the Dicotyledons, Vol. 2.
Thoday, D., and Johnson, E. T., 1930. On Arceuthobium pusillum, Peck II, Flowers and Fruit, Ann. Bot., 44, p. 813.
Treub, M., 1882–3. Observations sur les Loranthaceae, Ann. du Jard. Bot. de Buitenzorg, II and III.
Van Tieghem, P., 1869. Anatomie des Fleurs et du Fruit de Gui, Ann. des. Sci. Nat., Tome 11–12, p. 101.
Description of Plates
Fig. 1.—Habit drawing Korthalsella Lindsayi parasitic on Helichrysum glomeratum. Natural size.
Fig. 2.—Habit of young plant of K. salicornioides parasitic on Leptospermum ericoides. Natural size.
Figs. 3 & 4.—Seedlings of K. Lindsayi on K. Lindsayi. Natural size; s.: seedling; note also germinating seeds.
Fig. 5.—Branch of K. Lindsayi with inflorescences in fruit. X 4.
Fig. 6.—K. salicornioides shoot in fruit. X 4.
Fig. 7.—Habit of K. clavata. Natural size.
Fig. 8.—K. salicornioides tip of shoot in flower; ♂: male flower; ♀: tip of female flower; poll.: exuding pollen. X 10.
Figs. 9 & 10.—Inflorescences K. Lindsayi. X 10.
Fig. 11.—Diagram of a longitudinal section of a K. salicornioides shoot with flowers; ♂: male flower; ♀: female flower; col.: collar; node: node. X 5.
Fig. 12.—Diagram of a longitudinal section of a K. Lindsayi inflorescence. X 5.
Fig. 13.—Diagram of transverse section of K. Lindsayi internode; gr. res.: groups of resin cells; ph. t.: photosynthetic tissue; v. b.: vascular bundle; cort.: cortex; ep.: epidermis.
Fig. 14.—Diagram of a transverse section of a K. salicornioides internode.
Fig. 15.—K. Lindsayi; a: female flower; b: male flower. X 30.
Fig. 16.—Fruits (a) K. salicornioides; (b) K. Lindsayi. X 10.
Fig. 17.—Germinating seed of K. salicornioides. X 12.
Fig. 18.—K. salicornioides seedling on K. salicornioides. X 10.
Fig. 19.—Section of germinating seed of K. Lindsayi; end.: endosperm collapsed, the food store largely withdrawn from it; rad.: radicle of seedling; cot.: cotyledonary collar still embedded in endosperm; g.: growing apex. X 110.
Fig. 20.—Longit. median section female flower of K. Salicornioides; st.: stigma; per.: perianth. X 110.
Fig. 21.—Longit. median section K. Lindsayi developing fruit in young stage; z.: fertilised egg; end.: segments of endosperm as yet without cell walls; v.: cells which are going to form the viscous layer becoming meristemmatic; st.: stigma; per.: perianth lobes. X 110.
Fig. 22.—Young embryo sac K. Lindsayi. X 235.
Fig. 23.—Spore tetrad in K. Lindsayi female flower; note faint nuclei. X 235.
Fig. 24.—Embryo sac K. salicornioides. X 235.
Fig. 25.—Longit. median section K. Lindsayi female flower. X 110.
Fig. 26.—Longit. median section K. Lindsayi developing fruit; emb.: embryo; end.: endosperm; v.: developing viscous layer. X 110.
Fig. 27.—Transverse section male flower K. salicornioides, slightly oblique; per.: perianth; pol. gr.: pollen grain; sep.: remains of septa between pollen chambers; f.: fibrous layer; ep.: remains of epidermis seen in places. X 110.
Fig. 28.—Longit. tangential section unopened male flower K. Lindsayi; per.: perianth; pol. gr.: pollen grain; tap.: tapetum; f.: fibrous layer. X 110.
Fig. 29.—Longit. median section male flower of K. salicornioides; per.: perianth; pol. gr.: pollen grain; pol. sac w.: pollen sac wall. X 110.
Fig. 30.—Longit. median section male flower K. Lindayi; pore: central pore; pol. gr.: pollen grain; f.: fibrous layer; ep.: outer layer of pollen sac wall; i.: remains of inner layer of wall; per.: perianth lobe. X 110.
Fig. 31.—Longit. median section K. salicornioides developing fruit; emb.: embryo; end.: endosperm; v.: viscous tissue; st.: stigma; per.: perianth lobe. X 110.
Fig. 32.—Longit. median section of mature fruit of K. Lindsayi; st.: stigma; per.: perianth lobe; v.: viscous tissue somewhat collapsed in prepara tion of section; emb.: embryo; end.: endosperm; cap.: parenchyma; cap; w.: wall cells flattened by internal pressure. X 110.
Fig. 33.—Transverse section of Leptospermum twig with K. salicornioides haustorium surrounding stele of host; p.: pith of host stele; w. v.: wood vessels; w. f.: wood fibres; h.: haustorial tissue showing considerable secondary growth. X 45.
Fig. 34.—Detail of young K. salicornioides haustorium in Leptospermum. X 425.
Fig. 35.—Transverse section of a vascular bundle in K. Lindsayi internode. Cellulose walls black, lignifled walls cross-hatched. X 220.
Fig. 36.—Transverse section of K. Lindsayi stem immediately above a node showing a young haustorium; h.: partly surrounding stele of host stem (st.); lat. br.: lateral branch bud; 1st col., 2nd col.: two successive collars of the bud; col.: tip of collar of the lower node of host stem. X 50.
Fig. 37.—Longit. section of K. Lindsayi stem with a seedling K. Lindsayi haustorium. Cell contents are stippled in the haustorial cells. X 90.
Fig. 38.—Transverse section part of a K. Lindsayi internode showing the young haustorium of a K. Lindsayi seedling. Nuclei are drawn in the haustorial cells; c.v. and s.v.: vascular elements belonging to the host; f.: fibre of host separated from the vascular elements by secondary growth of the haustorium at the point marked 2°. X 90.
Fig. 1.—Habit drawing Korthalsclla Lindsayi parasitic on Helichrysum glomcratum. Natural size. Fig. 2.—Habit of young plant of K. salicornionds parasitic on Leptospermum ericoides. Natural size. Figs. 3 and 4.—Seedlings of K. Lindsayi on K. Lindsayi. Natural size. s.; seedling. Note also germinating seeds. Fig. 5.—Branch of K. Lindsayi with inflorescences in fruit X 4. Fig. 6.—K. salicormoidcs shoot in fruit. X 4. Fig. 7.—Habit of K. clavata. Natural size.
Fig. 8.—K salicornioides tip of shoot in flower; ♂. male flower; ♀. tip of female flower; poll.: exuding pollen X 10 Figs. 9 and 10.—Inflorescences K. Lindsayi X 10. Fig. 11 — Diagram of a longitudinal section of a K. salicornioides shoot with flowers; ♂ male flower; ♀: female flower, col: collar; node node. X 5 Fig 12—Diagram of a longitudinal section of a K Lindsayi inflorescence. X 5 Fig. 13.—Diagram of transverse section of K Lindsayi internode; gr res. groups of resin cells; ph t.: photosynthetic tissue; v. b: vascular bundle; cort. cortex, ep.: epidermis. Fig. 14—Diagram of a transverse section of a K salicornioides internode Fig. 15—K. Lindsayi; a: female flower; b: male flower. X 30. Fig. 16.—Fruits (a) K. salicornioides; (b) K. Lindsayi. X 10. Fig. 17.—Germinating seed of K. salicornioides X 12. Fig. 18—K. salicornioides seedling on K. salicorniodes. X 10. Fig. 19.—Section of germinating seed of K Lindsayi; end endosperm collapsed, the food store largely withdrawn from it; rad radicle of seedling; cot.: cotyledonary collar still embedded in endosperm; g.: growing apex X 110.
Fig. 20.—Longit. median section female flower of K. salicornioides; st.: stigma; per.: perianth. X 110. Fig. 21—Longit median section K. Lindsayi developing fruit in young stage; z.: fertilised egg; end.: segments of endosperm as yet without cell walls; v: cells which are going to form the viscous layer becoming meristemmatic; st: stigma; per.: perianth lobes. X 110. Fig. 22.—Young embryo sac K. Lindsayi X 235. Fig. 23.—Spore tetrad in K. Lindsayi female flower; note faint nuclei. X 235 Fig. 24.—Embryo sac K. salicornioides. X 235. Fig. 25.—Longit. median section K. Lindsayi female flower. X 110. Fig. 26.— Longit. median section K. Lindsayi developing fruit, emb: embryo; end: endosperm; v.: developing viscous layer. X 110.
Fig. 27.—Transverse section male flower K. salicornioides, slightly oblique; per.: perianth; pol. gr.: pollen grain; sep.: remains of septa between pollen chambers; f: fibrous layer; ep.: remains of epidermis seen in places. X 110. Fig. 28.— Longit tangential section unopened male flower K. Lindsayi; per.: perianth; pol. gr.: pollen grain; tap.: tapetum; f.: fibrous layer. X 110. Fig. 29.— Longit median section male flower of K. salicornioides; per.: perianth; pol gr.: pollen grain, pol sac w: pollen sac wall. X 110. Fig. 30.—Longit. median section male flower K. Lindsayi; pore: central pore; pol. gr.: pollen grain; t: fibrous layer; ep: outer layer of pollen sac wall; i. remains of inner layer of wall. per: perianth lobe X 110.
Fig. 31.—Longit median section K. salicornioides developing fruit; emb.: embryo; endosperm; V.: viscous tissue; st. stigma; per. perianth lobe. X 110.
Fig. 33.—Transverse section of Leptospermum twig with K. salicornioides haustorium surrounding stele of host; p.: pith of host stele; w. v.: wood vessels; w. f.: wood fibres; h. haustorial tissue showing considerable secondary growth. X 45. Fig. 34.—Detall of young K. salicormoides haustorium in Leptospermum. X 425.
Fig. 36.—Transverse section of K. Lindsayi stem immediately above a node showing a young haustorum; h.: partly surrounding stele of host stem (st.), lat br.: lateral branch bud; 1st col. 2nd col.: two successive collars of the bud; col: tip of collar of the lower node of host stem. X 50.
Fig. 37.—Longit section of K. Lindsayi stem with a seedling K. Lindsayi haustorium. Cell contents are stippled in the haustorial cells. X 90
Fig. 38—Transverse section part of a K. Lindsayi internode showing the young haustorium of a K. Lindsayi seeding. Nuclei are drawn in the naustorial cells; c.v. and s.v.: vascular elements belonging to the host; f.: fibre of host separated from the vascular elements by secondary growth of the haustorium at the point marked 2°. X 90.