Amendments of the Limits of the Genera Gyromitra and Pseudorhizina, with the Description of a New Species, Gyromitra Montana

Home , Gyromitra californica, Gyromitra gigas, Helvellaceae, Helvella acetabulum, Rhizinaceae

Amendments of the limits of the genera Gyromitra and Pseudorhizina, with the description of a new species, Gyromitra montana

Harri Harmaja

Department of Botany, Uni ersity of Helsinki, SF-00170 H elsinki, Finland

\HARMAJA, H. 1973: Amendments of the limits of the genera Gywmitra and Pseudo rhizina, with the descr~ption of a new species, Gyromi tra montana. - Karstenia 13 : 48--58. - Some new observations, relevant for specific, subgeneric and generic taxonomy, are presented on the characters of the species of Gyromitra Fr. and Pseudorhizina J ac. (Disco mycetes, Pezizales), and the taxonomic significance of earlier known diagnostic features is dealt with and partly re-evaluated . On the basis of this information, it is considered that Pseudorhizina is a valid, independent genus, and that besides the type species, P. sphaero spora (Peck) Pouz., -it should also contain Gyromitra californica ( Phil!.) Rait . A tabulation comparing Gyromitra and Pseudorhizina. is presented. It is also emphasized that, as pre i ously suggested by the author, the genera Discina (Fr.) Fr. and Neogyromitra Imai should be included in Gyromitra, together forming a subgenus of their own, Gyromitra subg. Discina. Gyromitra gigas sensu McKnight is considered different from the true G. gigas, and is described as a new species, Gyromitra montana Harmaja. Seven new combinations, necessary due to the amendments of the two genera, are made: Gyromitra subgenus Discina (Fr. ) Hai'IIlaja, Gyromitra apiculatula. (McKn.) Harmaja, Gyromitra korfii (Raitv.) Har maja, Gyromitra macrospora (Bub.) Harmaja, Gyromitra olympiana (Kan.) Harmaja, Gyromitra warnei (Pecl: ) Harmaja and Pseudorhizina ca.lifornica ( Phil!. ) Harmaja.

I. Pseudorhizina Jac . In my earlier paper on the delimitation of species is indicated also by their macroscopic the genera of the family Rhizinaceae Bon. features. Having observed some additional (HA RMAJA 1969a) I included the genus Pseu features separating C. sphaerospora and C. dorhizina Jac. in the genus Cyromitra Fr. californica from the rest of the Cyromitra (nom. conserv.) . However, at the same time species, and re-evaluated the earlier known I reported, evidently for the first time, that facts, I now consider that there is justifica the type species of Pseudorhizina, \ iz. Cyro tion for accepting the genus Pseudorhizina, mitra sphaerospora (Peck) Sacc., and C . in order to accommodate . these two species. calif arnica ( Phill. ) Raitv. differed from the At present this genus is monotypic a nd is rest of the species of Cyromitra (sensu latis retained (e.g., EcKBLAD 1968 ) only because simo) in having no cyanophilic perispore of the spherical shape of the spores of its periplasm complex outside their true spore type species, but, as is shown below, I cannot wall. That structure is lacking during all accord this character the slightest diagnostic stages of spore development in these two value at the generic level! species, while in the rest of the species it gets Modern taxonomists specialized in the developed towards spore maturity, first cup-fungi ha e usually placed P. sphaero appearing at the spore ends. I also mentioned spora (Peck) Pouz. and C. californica in that a close relationship between these two different genera, though the striking resem-

48 blances in their macroscopic and microscopic hidden by the conspicuous de Bary bubble features (some of the latter were then still (in P. sphaerospora they may be observed unknown) were pointed out by SMITH (1949: in many immature spores and occasional 153, sub Helvella ) in North America, where mature ones devoid of de Bary bubbles), are both species occur. Even the striking red observed as a fairly distinct guttule at each staining of the stipe base, also reported in end of the spores in G. californica but may some European collections of P. sphaerospora be represented by only one larger drop in (e.g., PouzAR 1961, RAHM 1970 ), is an almost P. sphaerospora. In both species the position constant feature in both species. PouzAR of the largest drop or drops is usually not ( 1961) gives the status of forma to the red symmetrical (see Fig. 1: a, b in H ARMAJA stiped ascocarps of P. sphaerospora, but no 1969a). One other difference between the thing will be gained through such a nomencla species is the much wider di~tribution of P. tural procedure, especially as it has not been sphaerospora, G. californica being restricted proved that the presence/absence of the red to northwestern North America. In addition, colour would depend on genetic factors. In Dr. SMITH, who has field experience of both deed, I do not know at present of any other species, writes (SMITH 1949 ) that P. sphaero morphological differences between the two spora usually grows on decaying wood, while species than a few sporal ones. the substrate of G. californica is more or less The spores of P. sphaerospora are com moist soil, either disturbed or undisturbed. pletely globose and rather small, being 8.0- Apart from possible minor differences in the 12 .0 [liD in diameter, and the half-mature anatomical features of the fruit body, a slight and mature spores, at least in dried speci phenological difference might possibly be mens, each possess one central to slightly ec revealed if extensive material of both species centric de Bary bubble (visible at least in from climatically similar areas were available water, 5 % KOH, Melzer's reagent. and cotton far comparison (does P. sphaerospora gene blue; I do not know whether these de Bary rally occur somewhat later in the spring?) . bubbles, already observed by me before the For the reasons presented above, I consider publication of my earlier paper but not H elvella sphaerospora Peck and H elvella ca mentioned in it, have previously been report lifornica Phill. congeneric, being the only ed in these species). In G. californica the two species belonging to the genus Pseudo spore shape is elliptical, slightly inequilateral; rhizina ]ac., which accordingly is considered the spores are larger, ca. 13.5- 17 .5x7.5- valid. 9.5 [liD, and only a very small proportion In Table 1. I have summarized the .most contain one de Bary bubble towards maturity, important diagnostic differences between the mostly asymmetrically located. As a result of redefined genera Pseudorhizina and Gyromit the different shapes of the spores of the two ra (the distinctions between Pseudorhizina species, the inconspicuous sporal oil drops, and the two genera H elvella and Rhizina when present at all and visible without being Fr. being more readily apparent) .

II. Gyromitra Fr. Recently there has been a general trend BENEDIX ( 1969) has adopted a completely among taxonomists to reduce the number of different approach, splitting genera exten the genera of the family Rhizinaceae (e.g., sively, but unfortunately some of his solutions EcKBLAD 1968, HARMAJA 1969a, RAITVIIR appear to be extremely artificial, e.g. the 1970, McKNIGHT 1971, SvRCEK & MoRAVEC location of the species perlata and leuco 1972, KoRF 1972 ). While mostofthese authors xantha in different genera, and the attri amalgamate t!he genus N eogyromitra Imai bution of Discina and Rhizina to different with the genus Discina (Fr.) Fr., M cKNIGHT families. DISSING ( 1972 ) also prefers narrow reduces the number of genera in a different generic concepts and keeps Discina and Neo way; he unites the Gyromitra and Neogy gyromitra apart. romitra species because of their distinct pileus I am still of the opinion that Discina and and stipe in Gyromitra, but keeps Discina Neogyromitra should not only be joined apart, in irs classical concept, i.e. accommo with each other, but also be merged in Gy dating discoid, indistinctly stipitate species. romitra, which becomes much more homo-

49 Table 1. A comparison of the most important characters separating the genera Pseudorhizina Jac. and Gyromitra Fr.

Pseudorhizina Gyromitra

1. Underside of the »cap»l very finely whi t~s h granulose or glabrous as seen by the naked eye tornentose

2. Stipe with very prominent, high, sharp- smooth or with less conspicuous, low, edged ribs continuing far outwards on more or less blunt-edged ribs the underside of the cap, resembling those of H elvella acetabulum

3. Stipe base in fresh ma-1 very often strikingly red never with red colouration ture fruit bodies

4 . Length of most mature under 17 fA-ill over 17 fA-ill spores (excluding any perispore)

5. Cyanophilic perispore-1 absent I present periplasrn complex

6. Contents of unripe moderately cyanophilic cyanophobic to weakly cyanophilic spores, the oil drops ex- eluded (before any perispore-periplasm has sur- rounded the spores)

7. Oil drops in spores from not always distinct; when present their always present, with a fixed sym- dried fruit bodies position very often asymmetrical, their metrical position and size, usually also number and size also being more or constant number within species less variable even within species

8. De Bary bubbles in present towards maturity, either in a!- absent spores from dried fruit most all spores, or m a small pro- bodies portion

9. Paraphyses in KOH fading to more or less hyaline in a few distinctly coloured, fading more slow- (e.g. , 5 %) minutes, their contents sometimes re- ly, either red-brown due to encrusted taining some of the original colour and/or more or less homogeneous in- and then being very pale greyish tracellular pigment, or with more or brown, and homogeneous less granular yellowish-brown contents

10. Excilpular tissue of ap- not reviving easily, many hyphae re- reviving more or less easily, hyphae propriately dried fruit rnaining collapsed, perhaps at least readily becoming inflated, having bodies as observed m partly because they seem to have so- more or less thin walls heated cotton blue rnewhat thickened walls

11. Cyanophilic »septal conspicuous, being common and of scattered at most, often practically belts» (seep. 55) on ex- variable breadth lacking, always very narrow and in- cipular hyphae conspicuous geneous when the species sphaerospora and 1. Shape of the fruit body californica are removed (see above). With the aid of some new observations, I shall As pointed out earlier (HARMAJA 1969a), the distinction between subsessile, shortly attempt to clarify my earlier criticism of the stipitate and distinctly stipitate species is four features which at present are the main very vague. This is evident also on p. 46 of criteria of those wanting to retain Discina or McKNIGHT's paper ( 1971 ), though that both Discina and Neogyromitra. author still considers Discina as an indepen dent genus. These differences, often used at the generic level, are quantitative only, de-

50 pending on the degree to which the stipe cause of great similarities in other features has developed. All the species concerned considered to have more diagnostic value at seem to have at least the beginning of a the generic level, especially the anatomy of stipe, and it is well known in the literature the excipulum and the spore characters. A that, e.g., G. per lata (Fr.) Harmaja of the striking example showing how careful one Discina group may posses a distinct stipe and should be when weighting the value of asco not rarely even a gyromitroid hymenium, i.e. carp shape at the generic level, is the well a conspicuously folded (in age it is as a rule known case of Peziza proteana with its spa at least slightly folded) and somewhat con rassioid form, which repeatedly had given rise vex one. Also the pictures of BENEDIX ( 1969) to a new genus! The principle followed e.g. representing D. melaleuca Bres. and Para in the delimitation of the three genera men discina intermedia Bened., likewise of ·the tioned above, should also be adopted in the Discina group, show fungi fairly gyromitroid present case, when, as I believe, it has indeed in appearance. On the other hand, McKNIGHT been proved that those microscopic features ( 1971: 46 ) reports having found a fruit suggest a close relationship between the spe body of D. korfii Raitv. (as G. fastigiata cies of Gyromitra s. lato. One single, vague [Krombh.] Rehm) of the Neogyromitra macroscopic quantitative character simply group with an entirely discinoid hymenium. does not justify the distinction of genera. The convex pileate, gyromitroid to lobate, EcKBLAD ( 1968 : 132) came to the same or helvelloid, hymenium with an edge free at conclusion as regards the Pezizales generally. first and sometimes even later, too (in G. This is also the trend in the taxonomy of, say, tasmanica Berk. & Cooke rarely also in G. the Agaricales and the Aphyllophorales. ambigua [Karst.] Harmaja), and the di·s cinoid one, as well as the small stipe and the 2. Apiculi of the spores long and/or thick stipe obviously only repre sent different stages in the same evolutionary As LE GAL ( 1947) has so excellently de series where the specialization of the fertile monstrated, the apiculi of Discina and N eo and sterile parts of the apothecium into a gyromitra spores are swellings at the spore very distinct pileus and stipe and/or vice ends in a perispore, the spaces between the versa is in progress. The hymenium type is perispore and the spore wall proper being correlated with the size of the stipe, the hy filled with periplasm. Both the perispore and menium being convex and mostly also at the periplasm are distinctly cyanophilic. I, tached to the stipe when the stipe is pro in turn, have shown (HARMAJA 1969a) that minent and well developed. exactly the same structure is present in the Very important is that the anatomy of the spores of Gyromitra s. str. The much used more developed stipes in the esculenta-infula attribute »apiculate» is, however, still em and N eogyromitra groups is not different ployed to designate a supposed qualitative from that of the small Discina stipe, these diagnostic difference between the »apiculate» anatomical characters being always also iden spores of N eogyromitra and Discina and the tical with those of the excipulum. This fact »non-apiculate» ones of Gyromitra s. str. The confirms the opinion about a close relationship differences concerned are, of course, only between the prominently stipitate and shortly quantitative, depending on the different ex stipitate groups. Moreover, it becomes un tent of the apical inflation of the perispore in masked that in all groups concerned this different species. The difference »apiculate» sterile part connecting the fruit body to the versus »non-apiculate» is thus completely substrate is only a very weakly specialized worthless at the generic level, especially as organ, being only an elongation of variable it has now been clearly demonstrated . that length of the connecting point composed there are species of the Neogyromitra and of the same kind of tissue ( textura intricata) Discina groups that almost or completely as the outer layer, or excipulum, of the fer lack »apiculi», namely G . gigas (Krombh.) tile part of the fruit body. Que!. &ensu McKNIGHT 1971 (which will be According to modern taxonomy, genera described as a new species, G . montana Har such as H elvella, Otidea and Sarcoscypha maja, later in this paper), D. apiculatula comprise subsessile to distinctly stipitate spe McKn., D. olympiana Kan., and ob iously cies, which are treated as congeneric be- (according to the text and photo in BENEDIX

51 1969) also D. melaleuca. In other words, in Discina sensu EcKBLAD 1968) . Towards ma these species the perispore is only slightly turity the perispore in the latter group be more detached from the spore wall at the comes distinctly to faintly rough, always with spore ends than it is elsewhere on the spore, a reticulate pattern under the light micro~ often being even less inflated than in some scope, evident at least at high magnifications. species of Gyromitra s. str., especially G. am In apiculi this ornamentation is more delicate bigua! In addition, I have found that in few than elsewhere. In Gyromitra s. str. the pe scattered spores of G. ( s. str.) tasmanica (of rispore appears smooth. Even if this diffe which two specimens from New Zealand rence were supposed to be constant, as it at and Chile were kindly presented to our first appears to be, I should not wish to treat herbarium by Dr. H. DrssrNG, Copenhagen) it as one of the most essential diagnostic the perispore had one to several papilla-like generic characters. It should be noted that inflations, especially near and at the spore many authors who keep the esculenta-infula ends, the appearance of these spores thus group and the Discina-Neogyromitra group being very much the same as that of the in different genera, at the same time accept spores of G. caroliniana (Fr. ) Fr. of the elsewhere in Pezizales the inclus·ion of both N eogyromitra group. smooth and rough and even very clearly In this connection it may be mentioned ornamented spores in one genus (e.g., Chei that I have observed a moderately cyanophi lymenia, Lamprospora, Octospora, Peziza and lic perispore in both mature and immature Pulparia). Moreover, the reticulum is very spores of Microstoma protractum, periplasm faint in some species of the Discina-N eogyro being, however, not discernible. In most spo mitra group, namely G. gigas sensu Mc res, both ripe and immature, this perispore KNIGHT, D . olympiana and D. apiculatula, is everywhere smooth and tightly appressed many mature spores in these species even ap to the true spore wall, thus not being easily pearing smooth unless studied very carefully. observed. However, in a certain proportion On the other hand, when I examined ma of the spores it is detached at the spore ends, terial of G. ambigua, a species of the escu forming distinct, more or less regular ap]culri. lenta-infula group whose spores are least Small papilla-like swellings in the perispore, dissimilar to those of some species of the especially near the ~pore ends, like those des Disci na-Neogyromitra group (e.g. G. gigas), cribed a:bove in G. tasmanica, may often a:c I observed that some aberrant spores had a company these apiculi, or occur alone. In distinct and regular reticulum, also observ other words, in Microstoma protractum, the able as a roughness on the sides of the spo same hymenium, or even the same ascus, may res. After this observation I studied the sur contain spores of »Gyromitra type», »Neogy face of normal G. ambigua spores very care romitra gigas or Discina type» and »Neogyro fully, and it did not seem impossible that, mitra caroliniana type»! Similar examples of contrary to present opinion, at lea:>t a certain the coexi~tence of different spore »types» can proportion of them may possess an extremely easily be found elsewhere in Pezizales (see faintly reticulate perispore. Studies of the e.g. LE GAL 1947). It seems that when a perispore with both the scanning and the perispore is present in Pezizales, it has fre transmission electron microscope would be quently a tendency to become inflated, either most important to ascertain whether it really constantly or occasionally, especially at the as a rule is ornamented in G. ambigua (and spore ends when the spore is elongated, the rest of the species of the esculenta-infula the resultant structures, or inflations being group). If this proved to be a fact, the pre regular or irregular in shape, the inflation sent character could definitely not be consi pattern as a whole being however, constant dered at the generic level any more, the dif to a high degree within species. ferences observed only being quantitative in that case. 3. Reticulum on the spores In Microstoma protractum I also observed the perispore to be more or less clearly re The surface of the spores provides a well ticulate, but only in those spores where the known, possible true, difference between the perispore also displayed swellings, apiculi esculenta-infula group ( = Gyromitra s. str.) and/or papillae (see above) . This observ and the Discina-Neogyromitra group ( = ation suggests that in Microstoma the form-

52 ation of swellings in the perispore and the reticulate perispore loses muoh of its dia presence of a reticulum (as seen from abo e gnos-tic value in taxonomy. However, in Rhi under a light microscope) are positively zinaceae smoothness and roughness of the correlated with each other, in which case perispore is obviously genetically controlled they could not be treated as two independent and more constant within the species !!han characters. The swellings and the reticulum in Microstoma protractum, and accordingly may even be the same structure at a different this feature deserves recognition as an in stage of development and/ or iewed at diffe frageneric diagnostic character. Uneven rent angles, or two different structures re growth is well known in the hymenia of many sulting from one and the same phenomenon, species, especially G. esculenta, resulting in the evident tendency of the membranous pe folded, or gyrose h ymenia or pilei, but has rispore to grow unevenly in places and thus quite rightly long been considered to lack become inflated and folded here and there. validity as a generic character. As a perispore is always a very thin and fairly soft and elastic structure as compared 4. Number of oil drops in the spores with the spore wall proper, very slight ef fects, even non-genetic factors, may be suf Apart from the presence versus absence of ficient to alter its shape and make it apiculate the reticulum, the difference in the number and/ or ornamen1'ed. A comparison of the of oil drops (whose occurrence within the esculenta-infula group and the Discina species in question is very constant), is the Neogyromitra group reveals that in the for only apparently reliable criterion according mer group the perispore is smooth and only to which each species of Gyromitra s. lato very slightly inflated at the spore ends. In may be divided into one or other of two some forms of G. esculenta (Fr.) Fr. it even groups. The diffe•rences are well known, being lies at the same distance from the spore as follows: In the esculenta-infula group there wall throughout. In the Discina-Neogyromit are two equal-sized oil drops symmetrically ra group, with its generally larger spores, the located on both sides of the median plane perispore is rough and adorned with reticu of the spores, while in the rest of the species, lum, and the apical swellings of the perispo i.e. the Discina-Neogyromitra group, there is re are generally larger and more clearly de an odd number of oil drops, either (most fined. Three exceptional species with indis commonly) three guttules with a larger one tinct apiculi, G. gigas sensu McKNIGHT, in the middle of the spore and two smaller D. olympiana and D. apiculatula, mentioned ones situated symmetrically on both sides of above in section 2, are each very closely the large one near the spore ends, or a single related to certain distinctly apiculate species, large central drop. Some species (D. mela and moreover, all three have fainter reticuli leuca and Paradiscina intermedia, BENEDIX than the latter (compare with the observ 1969 ) which evidently belong here (the ations on M icrostoma described above! ). structure of their excipulum is unfortunately The situation seems to be similar with the not adequately known) seem to have a good fourth of these »exceptional» species, D . m e proportion of biguttulate spores, but the laleuca. On the other hand, G. ambigua has drops are then both of unequal (rarely equal) the widest apiculi in the esculenta-infula size and asymmetrically located in a way group and is also the species with a tendency which clearly suggests that they are derived to develop a reticulum (see above) . from three original drops, the larger of the Until studies with the scanning or perhaps two having been formed by the merging of a preferably the transmission electron micro small lateral drop in a large central one. This scope are available, we can suppose that the situation has been excellently described and reticulum may originate in at least two diffe illustrated by BENEDIX ( 1969) . rent ways: either it is a distinct substance However, I do not consider the difference accumulating on the perispore in the form of in the number of oil drops in Gyromitra s. a reticulum, or it simply consists of minute str. and Discina sensu EcKBLAD (1968) dia swellings and folds in the perispore. In the gnostic at the generic level, since it is a purely latter case, which would accord with the quantitative difference, clearly depending on apparent situation in Microstoma protractum differences in spore size and shape. Some described above, the difference smooth versus other genera contain species with different

53 numbers of oil drops (e.g. H elvella and in these species the evolution towards the Leucoscypha) , others e en have species with evidently more suitable (for these small spo homogeneous or variably guttulate spore con res) biguttulate condi:tion is already in pro tents (e.g. Peziza), and modern taxonomists gress, judging from the descriptions and pho do not generally split these genera according tographs of BENEDIX ( 1969). to this character. It seems evident that in A similrur situation exists in, e.g., the genus many genera of Pezizales, Gyrontitra s. lata Helvella, where only H. macropus has large included, the oil drops, when present, have (for that genus) and fusiform spores, which a tendency to fill the space available inside are three-guttulate with the large drop in the spore, and to do this as economically as the middle. In addition, the spores of that possible. When the spore is globose, one drop species have very small warts, which, too, are is commonly present. When it is subglobose unique in H elvella (according to Drs SING & or elliptical to somewhat oblong, one large NANNFELDT 1966, SMITH WEBER 1972, and my elongated drop is (as usual in Helvella) or own observations). In spite of this, DrssrNG two drops are often a constant feature. When ( 1966) and most recent authors quite cor the spore is more or le~s fusiform, it is ea rectly do not consider this deviating species sy to understand that a three-guttulate pat worts a genus of its own (in fact, they do tern gives the most even distribution of oil not even regard it necessary to have a mono within the spore, the largest drop being lo typic section for this species!) . Why should cated in the middle, in the broadest point of we, contrary to H elvella, split Gyromitra with the spore. It also appears that the spores in regard to the number of oil drop~? many genera of Pezizales often tend to be globose when small, subglobose to elliptical Relations of Gyromitra s. str. and Discina and oblong when of moderate size, and sub sensu Eckblad fusiform to fusiform when large. These trends (there may also be an upper limit to The species of the .esculenta-infula group the diameter of an oil drop) may well ex are obviously very similar to those of the plain the fact that the smaller, mostly el Discina-N eogyromitra group in respect of liptical to oblong spores of the species of the habitus, colours, spore charG!!cters anatomy esculenta-infula group have two equal drops and ecology, as has already been pointed only (also those of G. ambigua, which, how out by me (HARMAJA 1969a). A couple of ever, are more or less fusiform; this species common features may be added. Under resembles, however, the Disci na-Neogyromitra ultra-violet light (the wave-length of the group more than the rest of the species of lamp being 254 nm) the dried fruit bod~es its group, and G. ambigua might perhaps of at least the following species were found be considered as a connecting link between to have more or less similar reactions in my the infula and N eogyromitra groups) while so far somewhat preliminary studies: G. escu the larger, more or less fusiform spores in lenta, G. tasmanica, G. infula (Fr.) Quel., the Discina-Neogyromitra group have the G. ambigua, G. gigas G. gigas sensu Mc three-dorp pattern or, infrequently, that KNIGHT, G. per lata, G. leucoxantha (Bres.) with one large central drop, which is ob Harmaja, D. apiculatula, D. korfii, D. mac viously derived· from the three-dropped one. rospora Bub., D . olympiana and D. warnei Rhizina undulata Fr. of the same family and (Peck) Sacc. (in addition, H elvella leucome same evolutionary sequence has long, narrow laena, G. californica, and P. sphaerospora, in spores with a tendency to partly parallel which only the hymenium and underside we walls. It has accordingly one more oil drop, re studied). These reactions were roughly as i.e. , four of them, two large ones in the follows: hymenium showing its natural colour middle part of the spore and two small ones, mixed with violet tints (at least in G. infula, each near the narrow erids. The small-spored G. gigas and D. warnei with faint olive tinge but three-guttulate D. melaleuca and Para in addition) ; underside as well as stipe and discina intermedia as described and depicted context pale, with mixed pale violet and by BENEDIX ( 1969) seem to be derived from yellow to orange-yellow tints (the yellow and some typical member of the Discina group orange-yellow tints indistinct in G. infula, and to have this number of drops as a relic G. ambigua and D . warnei; these bright co character. This is supported by the fact that lours generally tend to be more distinct in

54 specimens collected long ago, the ~iolet tinge can be seen, besides microscopically, often prevailing for a certain time after drying) ; also with the naked eye as the darkening of basal tomentum iolet. Rhizina undulata the medium near the cut surface against a seems to have different reactions since the light source or white background (e.g. paper) , underside of the fruit body appeared com provided that the section is sufficiently thick pletely dark violet, and a yellow zone was containing a large amount of paraphyses to seen in the upper part of the context. The provide the pigment enough. In the species reaction of its hymenium is similar to that of the esculenta-infula group the pigment of of the above species, showing a faint olive the paraphyses is likewise soluble in KOH tint, too. (perhaps slightly less easily? ), but it is usually Preliminary studies showed that curious darker being most evident as a reddish brown elements, cyanophilic »septal belts», were pre colour in the encrustments, and this colour sent in all the species examined in the is sometimes also present in the invariably esculenta-infula and Discina-N eogyromitra homogeneous contents of the paraphyses. groups, but were always infrequent and in SMITH ( 1949) noted that the contents of conspicuous, as was stated in the tabulation the paraphyses of »G. gigas» ( = obviously of the diagnostic characters of the genera essentially G. gigas sensu McKNIGHT ) are Pseudorhizina and Gyromitra. In G. escu yellow and granular in KOH, while EcKBLAD lenta and D. olympiana they occur occasio ( 1968 ) reports that the paraphyses of Dis nally; in other species, e.g. G. infula and G. cina pos•sess »brown or yellow granular con leucoxantha, only one or two of them were tents», failing, however, to tell the mountant. observed per anatomical preparation. They KANOUSE ( 1948 ) and SMITH ( 1949 ) also are also rare in Rhizina undulata (where, write that G. esculenta and G. infula have however, other cyanophilic matter occurs, e.g. red to reddish brown paraphy es in KOH. the contents of certain hyphae), but, as staot:ed All these observations suggest tha:t there in the comparative tabulation, they were might be a true difference in the pigment frequent and conspicuous in the genus (or pigments) of the paraphyses, and it is Pseudorhizina. These structures are simply accordingly hoped that the constancy of this strongly cyanophilic matter surrounding the difference would be tested by those studying hyphae of the excipulum just at the septa these fungi. The same applies to the in the form of a ring. Sometimes, such as in occurrences of red (not reddish brown! ), Gyromitra s. lato, they seem to be no bmader often thread-like particles (some connection than the thickness of the septal wall, i.e., with DNA? ) which I observed in the asci they are extPemely narrow; in other cases of G. esculenta, G. tasmanica, G. infula and they are bmader, extending equally far along G. ambigua, outside the developing spores, the hypha on both sides of the septa. They when they were mounted in 5% KOH, while never seem to occur around all the septa they were absent from the asci of the species but only a proportion of them. (I have also of the Discina-Neogyromitra group. However, found them in e.g. Tarzetta [Pustularia] ca if these two chemical differences prove to tinus, Otidea onotica, 0. indivisa, 0 . phle be valid, their significance should not be bophora, 0 . cf. leporina, 0. spp. indet., overemphasized. Chemical differences are H elvella acetabulum, H . lacunosa, Peziza ba commonly revealed, e.g. by the same KOH dia and Microstoma protractum.) treatment, in even very closely related species On the other hand, a couple of evident of the same genera in, for instance lichens slight differences between the esculenta and agarics. infula and Discina-Neogyromitra groups must In summary, there are two rather easily be reported. Though all the species of the observed ani constant differences between latter group have not yet been examined the species of the esculenta-infula group and in this respect, it appears that in KOH those of the Discina-Neogyromitra group, in (e.g., 5 %) the pigment of the paraphyses the characters of (I) the oil drops in the of these species is most evident in the con spores and (2) the surface of the perispore tents of their upper parts, being yellowish (it is, however, not sure whether this diffe brown and more or less granular. The pig rence is true even at the light microscope ment is soluble in this mounting medium, as level ; see the comment on G. ambigua in the

55 third section), and two additional ones, the subgeneric name Gyromitra subg. Discina validity of which has so far not been defini (Fr.) Harmaja, n. comb., is proposed for the tely proved, in the KOH reactions of (3) Discina-N eogyromitra group. »Gyromitra the paraphyses and those of (4) the extraspo subg. Lacunaria Fr.» (FRIES 1872: 174) is an ral contents of the asci. For the reasons stated invalid name, a nomen nudum, to be rejected earlier in this paper, I do not consider these since FRIES failed to supply any rlescr~ption differences sufficient for a generic distinction or diagnosis (International Code of Botanical but regard them as infrageneric, and as they Nomenclature: Art. 36) and only writes: seem to be correlated with each other, it »Hae duae species peculiare subgenus, La appears that the best solution is to consider cunaria dictum, forment.». The two species these two groups of species as two subgenera. referred to were commented earlier in his While the esculenta-infula group will con paper and are G. labyrinthica Fr., n. sp. ( = stitute Gyromitra subg. Gyromitra, G. escu G. gigas, as already judged by NANNFELDT, lenta being the type species of the genus, the 1932), and G. caroliniana.

III. New combinations, with notes on some taxa concerned Gyromitra subgenus Discina (Fr.) Harma 2). McKNIGHT ( 1971) treats for the most ja, n. comb. (Peziza A. Discina Fries, Systema part G . korfii under the name G. fastigiata mycologicum . . . 2: 38. 1822. - Lectotype (Krombh.) Rehm, but this is an obvious Peziza perlata Fr., selected indirectly by FRIES misapplication of that name, carefully proved himself [ 1849] through restricting to mono by SvRCEK and MoRAVEc (1972 r (The Euro typic the taxon which he at the same time pean specimens of G. fastigiata sensu Mc gave the generic statu . ) KNIGHT most probably represent the true Gyromitra apiculatula ( McKn. ) Harmaja G. gigas.) n. comb. (Discina apiculatula McKnight, Gyromitra macrospora (Bub. ) Harmaja, Mycologia 61: 616. 1969. - Holotype [BPI] n. comb. (Discina macrospora Bubak, Ann. studied.) mycol. 2: 395. 1904. - Lectotype [BPI] Gyromitra korfii (Raitv.) Harmaja, n. studied.) - Besides the obvious spore diffe comb. (Discina korfii Rait\ iir, Tartu riikliku rences between this species and the closely iilikooli toimetised 268. Botaanika-alased related G. perlata, so thoroughly described t66d 9: 371. 1970. - Holotype and two by McKNIGHT ( 1969), I found also the tips paratypes [CUP] studied.) - In the original of the paraphyses different in these two description the only morphological difference species. At least in the type o.f G. macrospora given between this North American species they are somewhat thinner than those of G. and the very closely related G. gigas was the perlata, being only ca. 4.0- 8.0 flm in dia more slender spores. My observations suggest, meter (in Melzer's) . however, that the spores of G. korfii are also Gyromitra olympiana (Kan.) Harmaja, n. slightly shorter (which to be fair, is also comb. (Discina olympiana Kanouse, Myco apparent in RAITVIIR's spore diagram), the logia 39: 648. 1947.- Fragment of holotype reticulate ornamentation of the perispore is [MICH] studied.) slightly more delicate, and that its paraphyses Gyromitra warnei (Peck) Harmaja, n. expand more abruptly, even to a capitate comb. (Peziza warnei Peck, New York State head, which is broader, up to ca. 13 flm in Mus. Rep. 30: 59. 1878. - Part of holotype diameter (in Melzer's) . This average diffe [ NYS] studied.) rence in the paraphysis morphology is ana Pseudorhizina californica ( Phill.) Harma logical with that between G. montana and ja, n. comb. ( H elvella calif arnica Phillips, G. gigas (see below) and between G. Trans. Linnean Soc. II. Bot. 1: 423. 1880.) infula and G. ambigua (HARMAJA 1969b, Fig.

IV. Description of a new species, Gyromitra montana Harmaja Gyromitra montana Harmaja, n. sp. - culis sporarum dissimilibus et minoribus, V alde similis Gyromitrae gig antis, sed ab ea reticula perisporae leniter tenuiore, para differt forma sporarum plus ellipsoidea, api- physibus apicibus crassioribus, fructificatione

56 praecociore et distributione dissimili in re culatula is closely related to G. montana, gionibus montosis. - H olotype: »Gyromitra but is easy to recognize when it is kept in gigas (Krombh. ) Que!. I [U.S.A.,] Wyoming: mind that the former has a discinoid fruit West side of Teton Pass, / Teton Co. I body and thinner tips of the paraphyses. on soil around and under snowbanks, I co This new species is Gyromitra gigas sensu nifer forest. I Coll.: K. B. McKnight & J. B. SMITH ( 1949 ) and McKNIGHT (1971). How McKnight [VI.- VII. 1967] I Det.: K. H . ever, as G. gigas has originally been de M cKnight 10351» (H; isotypes K , BPI) . scribed on the basis of Bohemian material, This species, a specimen of which was it is exclusively that material that is to be kindly presented to our herbarium by Dr. consulted when interpreting and typifying K. H. McKNIGHT, differs from G. gigas this name. The current European meaning through, the perispore excluded, slightly mo of G . gigas, different from that of these Ame re ellipsoid, less fusiform spores with some rican authors, is correct and to be maintained what broader ends, the inconstancy of the as is obvious on the basis of e.g. a study of presence of the spore apiculi, the variable Czechoslovakian authors ( S VRCEK a MoRA VEC and often irregular shape and smaller size 19 7 2) . Also specimens now referable to G. of the latter when discernible, the slightly korfii have in North America been taken for more delicate ornamentation of the perispore, G. gigas, but it appears that in fact the oc the thicker tips of the paraphyses which may currence of the true G . gigas on that conti even be capitate and attain a breadth of ca. nent still remains to be proved, as also suppos 13 1-1m (as measured in Melzer's reagent) ed by RAITVIIR ( 1970) . while in G. gigas they are clavate to sub As the description of McKNIGHT ( 1971 , as capitate and only reach a diameter of ca. 10 G. gigas ) of G. montana, with photographs f!m, the earlier fruiting time often near melt and the map of its North American distribu ing snow (sometimes even developing fruit tion, is excellent and obviously based on bodies under the snow! ), and the different homogeneous material, I do not feel it im distribution in the mountains of western perative to prepare any from my side. How No-rth America and A ustria in Europe (the ever, some new observations on the charact fruiting time and the distribution according ers of this species are presented earlier in to McKNIGHT 1971) . From G. korfii this paper, including its comparison with the it can be distinguished through similar dif closest relatives. ferences as from G. gigas as regards the spo re shape, features of the spore apiculi, and Acknowledgements. -- I wish to express my the characters of occurrence, but also be sincere thanks to the following persons who most kindly provided herbarium material for ·this S>tudy: cause of the longer and broader ~pores of G. Dr. Henry Dissing (C ), Dr. John H. Haines montana. As compared with G. olympiana (NYS), Prof. Dr. Riohard P. Korf (CUP) , Dr. it is especially to be noted that the latter Paul L. Lentz (BPI), Dr. Kent H . McKnight has a discinoid ascocarp, longer and broader (BPI), D r. Robert L. Shaffer (MICH), and Dr. Stanley J. Smith (NYS). I am indebted also to spmes slightly more fusiform in shape, and Mrs. Anna A. Damstrom, M. A., for revising the thinner tips of the paraphyses. Also G. api- English of this paper.

REFERENCES BENEDIX, E. H . 1969: Art- und Gattungsgrenzen Discomycetes. A re-evaluation of their

57 HARMAJA, H . 1969a: A wider and more natural mitra gigas (Krombh.) Cke. - Friesia 1: concept of the genus Gyromitra Fr. - 34- 45. Karstenia 9: 9- 12. PouzAR, Z . 1961: Systematicka hodnota uchaoovce - »- 1969b: A neglected species, Gyromitra am sumavskeho - Helvellella gabretae (Ka bigua (Karst.) Harma:ja, n. comb., and G. vina) Pouz. et Svr. The taxonomical value infula s. str. in Fennoscandia. - Karstenia of Helvellella gabretae (Kavina) Pouz. et 9 : 13- 19. Svr. - Ceska Mykol. 15: 42- 45. KANOUSE, B. 1948: Some studies in the genus RAHM, E. 1970: Ueber einige Rhizinaceae aus dem Helvella. - Pap. Michigan Acad. Sci. Arts Hochtal von Arosa. - Schweiz. Zeitschr. Lett. 32: 83-90. Pilzk. 48: 77- 87 + I. KoRF, R. P. 1972: Synoptic key to the genera RAITVIIR, A. 1970: Once more on Neogyromitra of the Pezizales. - Mycologia 64: 937- caroliniana. - Tartu riikliku iilikooli toi 994. metised 268. Botaanika-alased tood 9: 364- LE GAL, M . 194 7: Recherches sur les ornement 373. ations sporales des Discomycetes opercules. SMITH, A. H . 1949: Mushrooms in their natural - Ann. Sci. Nat. XI: Bot. 8: 73- 297. habitats. - 626 pp. Portland. McKNIGHT, K . H. 1969: A note on Discina. - SMITH WEBER, N . 1972: The g-enus Helvella in Mycologia 61: 614- 630. Michigan. - Michigan Bot. 11: 14 7- '20 1. - »- 1971 : On two species of false morels (Gyro SvRCEK, M. a MoRAVEC, J. 1972: 0 druhu Hel mitra) in Utah. - Great Basin Nat. 31: vella fastigiata Krombholz. Ueber die Hel 35--47. vella fastigiata Krombholz. - Ceska Mykol. NANNFELDT, J. A. 1932: Bleka stenmurklan, Gyro- 26 : 1- 8 + I.