[Notes from Kew Type specimen, PRJ 2010] Kew images.

Fruitbodies of the type collection are small, slender, clavarioid plants, up to 1 cm. high, which are once or twice dichotomously branched. The branches are terete except toward the base of the sporophores and just beneath the dichotomies where they are somewhat flattened.
In other gatherings, robust, fastigiate specimens up to 7 cm. high occur, in which the lower branching may be polychotomous, and the subsequent dichotomies more numerous than in the type. It is sometimes difficult to trace the mode of branching owing to fusion of adjacent limbs. Fresh specimens vary in colour from white to dull pallid to flesh colour or deep buff, are of tough, fleshy consistency, and have a high water content (Crawford 1954). When dried these fungi become brown and horny-cartilaginous. Hyphal structure monomitic, consisting of generative hyphae 2-3-5 µ wide, with thin or very slightly thickened walls.
These hyphae bear clamps at the septa which may be of a more or less loop-like form. Hymenium thickening, reaching 156 µ on the main branches of some specimens, but not distinctly layered. Basidia 2 or 4-spored, tremellaceous. Spores hyaline, smooth, elliptical, but rather variable in both size and shape. Those of the type are 9-5-12 x 4-5-6 µ, but the complete range for the species is 9-13 (-16) x 4-5-6-5 (-8-5) µ. Spore germination as seen in the type specimen consists of the production of a short pointed sterigma about 5 µ long, from which a secondary elliptical spore 8 x 4 µ, is produced.

Distribution. This fungus has been reported under the various names listed above from: New South Wales, Victoria, Queensland, Australia ; New Zealand ; Tasmania ; Philippines ; Sumatra ; India ; Madagascar ; and Brazil.

Crawford (loc. cit.) wrote as follows on the genus Tremellodendropsis " Delimitation of the species of Tremellodendropsis is difficult ; morphologically they are all very similar, both in the fresh state and even more so when dried. Differences in size, length of stalk, degree of flattening of branches, and compactness of plant have been observed, but the growth form is most probably dependent on environmental conditions. In the related genus Aphelaria a similar wide range of growth forms has been recorded for A. dendroides. Spore size is usually regarded as a fairly constant character, but if the basidia produce from one to four spores a wide range in spore size from any one plant might be expected ". These observations are in complete agreement with those of the author. However, Crawford divided the genus into two subgenera on the basis of whether the basidia were cruciately subseptate at the apex, or whether they had a transverse septum cutting off a cruciately septate apical region before spore formation, and then used spore size and shape to distinguish the species. The present author has examined the type specimens of Clavaria pusio, G. flagelliformis and Thelephora archeri, and has found tremellaceous basidia of the kind figured by Crawford (fig. III, p. 624) for Tremellodendropsis transpusio in each. The septation is not visible in the very young basidia, and is not easily seen in the old ones which have shed spores, as they collapse readily in hymenial squashes. Even so in these old basidia there is nearly always a constriction at the point where the transverse septum would be expected, and it is.sometimes possible to see traces o£it projecting from the walls at this point. It is therefore often difficult to find basidia which clearly show complete septation, unless the fruitbody is in the correct stage of development, or unless suitable portions are examined. Nevertheless in some specimens of the type collection of C. pusio the basidia are well preserved and the septation easily seen. The description of T. transpusio by Crawford was thus quite unjustified, as apart from the structure of the basidium there is complete agreement in spore size (Crawford gave the spores of T. pusio as 9-16 x 4.5-7 µ and those of T. transpusio as 9-15 x 4-7 µ).
It has been found that there is exceptionally wide variation in spore size and shape in any one fruit-body. This is probably due in part to basidia producing from one to four spores as Crawford suggests, but in very few of the collections examined was there any close approach to the uniformity of spore size and shape as shown (fig. I, p. 620) for her various taxa. Indeed the spores of the type specimen of G. flagelliformis exhibited particularly wide variation in shape, with a high percentage showing the ovoid form of her var. ovalispora, and she herself states that the spores of this variety vary from subcylindrical to subglobose. There is considerable variation also in the size and shape of the spores of Thelephora archeri in which the majority are 9-12 x 4-5-6 µ, but a few are 11 x 7-5-8 µ. Spore measurements in these fungi are further complicated by spore germination with the production of smaller secondary spores on the parent plant. This may account for the lower range of spore size given by Crawford for T. flagelliformis, especially as she shows (fig. I (c), p. 620) a spore of T. flagelliformis var. tasmanica which has germinated to produce a typical pointed sterigma. With the exception of the var tasmanica, in which the spores appear to be somewhat larger, it seems. pointless to retain varieties based on these variable spore characters.
With regard to the type of Thelephora archeri it should be noted that Corner (1950) stated that "Lachnocladium archeri (Berk.) Bres. = Thelephora archeri Berk." and Cunningham (1953) wrote " The type labelled " Tasmania ", was referred by Bresadola to Lachnocladium; but Corner (1950, p. 723) Correctly showed it to be a Thelephora. A second collection on the type sheet ex " Delegate Hill, Vic,, E. Reader ". may be the same but this could not be ascertained since spores were not found ". It is difficult to see why these collections should have been referred back to genus Thelephora, as the elliptical spores are quite smooth, and hyaline or very faintly yellowish when mounted in potassium hydroxide, and the basidia are tremellaceous.
Crawford, discussing the systematic position of the genus Tremellodendropsis, writes as follows " Some authors would probably place Tremellodendropsis in the Tremellaceae, regarding the partial or complete septation of the apical region of the basidium as anomalous for the Clavariaceae; the subgenus Transeptia might even be placed in the Auriculariaceae because of the transverse septation. However, transverse septation of the old basidia occurs not only in members of the subgenus Tremellodendropsis but also in other members of the Clavariaceae-e.g., Clavulina it is therefore not unreasonable to suppose that the time of septation occurs earlier in the young basidia of the subgenus Transeptia. The members of the genera Aphelaria and Tremellodendropsis form a sequence from the nonseptate elongate clavate basidium, through the still e1ongate clavate but partially septate basidium, to the " Transeptia" basidium where the apical portion is cut off by a transverse septum to give a cruciately septate region. All these basidia differ from the typical Tremellaceous basidia in still being clavate not globular. The genus Tremellodendropsis provides a link between Clavariaceae, Tremellaceae and Aruiculariceae". In any consideration of the systematic position of these fungi the structure of the basidium assumes the greatest importance and so it is essential that a clear idea be formed of how the basidium develops. The clavate probasidium is cut off by a clamped basal septum, and continues to enlarge. A secondary septum which lacks a clamp connection is then formed cutting off an apical portion which becomes longitudinally septate and functions as the basidium proper, and a basal portion which functions as a stalk cell. Crawford considers that in the " Transeptia' type of structure described above (and according to the present author the only type found in Pseudotremellodendron) the term basidium should include both the longitudinally septate apical portion and the stalk cell. In the view of of the present author, however, the term basidium should be restricted to the apical segment, but it seems probable that this type of reproductive structure is a primitive one in the Tremellaceae. Crawfrd attaches great importance to the shape of the basidia in Tremellodendropsis and states that they " differ from typical Tremellaceous basidia in still being clavate, not globular". Whilst this is true, it must be remebered that the tremellaceous fungus Eichleriella spinulosa {Berk. and Curt.) Burt possesses basidia which are far more strongly clavate than those found in Pseudotremellodendron pusio. McGuire (1941) described two Sebacina (S. podlachica Bres., S. umbrina Rogers)'which have exactly comparable basidia to those described above. He writes of S. podlachia " probasidia at first clavate, with basal septa and clamps 5-15 µ below the swollen tips, tardily cut off by secondary septa at the bases of ths swollen tips, finally obovate........ etc. He adds "The great variation in basidium sizes cited by Bourdot and Galzin is probably due to the fact that the secondary basal septa are often very difficult to see. It may be that they sometimes fail even to develop,.... ", etc. Martin (1941) has also described similar basidia in Protohydnum cartilagineum A. Moll. he writes " The basidia are as Moller describes them, but he fails to emphasise sufficiently their unusual character. THe probasidia are at first long clavate (fig. 10), 30-35 µ in length. The terminal portion becomes greatly swollen and more or les globose, and finally is cut off from the basal stalk, the swollen portion only becoming divided by oblique or longitudinal septa". His figures, like those of McGuire, show no clamp at the septum between basidium and stalk cell. Again in Sebacina prolifera Rogers the basidia are rather similar to those of Pseudotremellodendron and Rogers (1936) writes of this fungus "It differs from all species of Sebacina known to the author in the broadened subbasidial cell". The only differences from the Pseudodotremellodendron basidium is that the septum between the basidium and the subbasidial cell is clamped, and the subsequent basidial proliferation occurs from this clamp connexion. This probably represents a more advanced type od reproductive structure in which the secondary septum of Pseudotremellodendron has been replaced by one that is primary and clamped. In al genera of the Clavariaceae where the basidium becomes transversely septate, this occurs after spore dischanrge, and is comparable with the fiormation of secondary septa in the vegetative hyphae, wheeras in these fungi it occurs before spore discharge, and forms part of the reproductive structure. In addition to the structure of the basidium, and nearly as important when deciding to which group of fungi Pseudotremellodendron should be assigned, is the method of spore germination. In no genus of the Clavariaceae does the spore germinate directly to produce a secondary spore, but this is a common occurence in the Tremellaceae. In this connexion it should be noted that Crawford figures (fig. I (c)) a germinating spore of T. flagelliformis var. tasmanica which has produced the characteristic pointed sterigma which in T. pusio has been shown to bear the secondary spore, and this is atxon which she maintains has clavate basidia which are only "cruciately subseptate at the apex". In view of this evidence the author considers that Pseudotremellodendron should be placed in the Tremellaceae. He cannot take seriously Crawford's suggestion of a possible relationship with the Auriculariaceae, from which this genus differs very widely.