| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sylvan Research, 198 Nolte Drive, Kittanning, Pennsylvania 16201
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSHISTORYRESULTS OF MOLECULAR AND...DISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
Agaricus brunnescens Peck sometimes is synonymized with A. bisporus ( J.E. Lange) Imbach, the cultivated button mushroom. However this relationship has not been accepted universally or unreservedly. For various reasons A. bisporus is the more commonly used name. This study demonstrated that A. brunnescens is based on a heterogeneous type, as suggested by elements of the type description, correspondence at NYS and a lack of consensus among studies of the type. DNA sequence analysis indicated that material corresponding to both A. bisporus and A. subrufescens Peck is present in the type as presently constituted. A lectotype corresponding to the concept of A. bisporus is chosen, and nomenclatural issues are discussed. A definitive history of A. brunnescens is attempted.
Key words: cultivated mushroom, lectotypification, nomenclature
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSHISTORYRESULTS OF MOLECULAR AND...DISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
, Isaacs (1967) and most particularly Malloch (1976) to be the correct binomial for the widely cultivated button mushroom, the basis of a multibillion-dollar global industry and the focus of diverse published research. However today the more widely used name for this species is A. bisporus ( J.E. Lange) Imbach, while even older disused names for the cultivated species exist (Redhead et al unpubl). Agaricus brunnescens has become controversial because, while Malloch and others (L. Parra pers comm) their own examinations have argued after that its type is conspecific with A. bisporus, Singer (1984) and Kerrigan (1987) reported different observations on (or interpretations of ) the type and respectively argued for, or suggested the possibility of, a different concept of Peck’s species.
One apparent mystery not previously discussed is why Peck and the several other principals in this story seem not to have considered the possibility that A. brunnescens might be the commonly cultivated species that we now call A. bisporus. Peck had discussed the cultivated button species twice before 1899 (Peck 1884, 1897). In the 1897 report FIGS. 8 and 9 depict "the Garden mushroom, Variety hortensis [of A. campestris; = A. campester in Peck’s usage]... often found in cultivation and in the markets, but... rarely found growing in the fields." True, in the tradition of the day the cultivated species was confounded with A. campestris L., a distinctly different species associated with grasses (see Redhead et al unpubl for a review of nomenclaturally significant examples). However even such an artificially broad species concept necessarily incorporated the range of variation established by inclusion of A. bisporus.
The cultivated mushroom had been a familiar species in the markets of Northeastern cities for at least 20 y by 1900. Falconer (1891) provided several relevant facts. During Reconstruction after the American Civil War the Southern states shipped increasing amounts of produce northward during winter, forcing Northern market gardeners to erect greenhouses to remain competitive. In some of these greenhouses a design that allowed cocultivation of salad greens and mushrooms was employed; florists subsequently might have begun to embrace this approach. Circa 1881 a purpose-built mushroom cellar 63 feet long was excavated and constructed at Dosoris, Long Island, New York. In 1891 mushroom spawn "[could] be obtained at any seed store." Falconer estimated that about 34 tons of European mushroom spawn was imported into Northeastern cities, principally New York, per growing season (the American spawn industry did not really begin production until after 1900). In 1891 a pound of mushrooms netted US$0.50 to the grower, about US$11.00 today (CPI method, Davies 2006).
Peck (1897) described outdoor cultivation in which "neglected places... are dug up and manure incorporated into the soil... Mushroom spawn... is then planted in the prepared places and a coat of fine manure spread over the surface." He also wrote that this species (A. campester in his usage) "is the one commonly cultivated and the one most often seen on the tables of the rich and of fashionable restaurants and public houses... [it is] perhaps more generally used and better known than any other... It is so eagerly sought in some of our cities that it is difficult to find the wild ones near these towns, for they are gathered almost as soon as they appear..." This 1897 description of an edible mushroom growing in manure and gathered at the earliest possible moment is evocative of the description (Peck 1900) of A. brunnescens as conceived by Peck in 1899 "... on deposits of manure and street scrapings... It is an edible species and is eagerly sought by Italians who are after them every day by 4 o’clock in the morning." Even Peck’s (1897) figures of A. campestris hortensis closely resemble his 1899 painting of the Noyes A. brunnescens specimens, which ostensibly constitute the type.
So Peck left us with a puzzle. Why would a familiar species with key features noted in 1897 not be recognized or even considered in 1899–1900, when almost identical features were recorded and incorporated into the type description of A. brunnescens? Perhaps the rarity (per Peck [1897], possibly drawing from European literature) of A. campestris hortensis in the field led him to discount this possibility. More plausibly, some features of (at least some part of) the material on which the A. brunnescens concept was based could have led Peck to his conclusion that a unique, novel species was in hand.
The question of A. brunnescens has remained open for a number of years and has complicated efforts to stabilize the nomenclature of the cultivated button species. One professional organization has published a recommendation for the use and conservation of A. bisporus over A. brunnescens (Edwards 1990). Such a conservation proposal, which also will establish the type of A. bisporus, has been drafted (Redhead et al unpubl). Thus a further attempt to finally resolve the typification and concept of A. brunnescens has become necessary.
Because the Noyes material (see below) as painted had the aspect of A. bisporus and material incorporated into the type has been determined confidently to be conspecific with A. bisporus (Malloch 1976, Parra pers comm), whereas my own observations differed (Kerrigan 1987), I eventually began to wonder whether the concept of A. brunnescens might have been based in part on additional material. Correspondence to Peck (at NYS) and Burt (at FH) concerning A. brunnescens strongly suggests that Peck’s concept of the species was a synthesis of data on different collections, provided by different collectors. Because only one specimen box, the NYS-designated type of A. brunnescens, and no other contemporary specimen mentioned in correspondence, was located at NYS, the further possibility that the type material might be a heterogeneous mixture of separate collections, possibly of multiple species, reflecting the apparently synthetic origins of the circumscription, could not be disregarded. Such a situation was documented for the type of A. californicus Peck (Kerrigan et al 2005), and the concept of A. pattersonae Peck presents a similar case (Kerrigan 1979).
To resolve this question the type material was reexamined in its entirety and the background correspondence linked to the relevant collections was reevaluated. DNA sequence analysis was brought to bear on assessment of the type of A. brunnescens for the first time.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSHISTORYRESULTS OF MOLECULAR AND...DISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
). DNA extractions were performed on 10 of the fragments.
|
Molecular analysis.— –
Preparation of cellular DNA via the CTAB method, amplification of ITS regions and cleaning of PCR products were performed as described by Kerrigan (2005). The primers of White et al (1990) are more clearly designated ITSp1 (vs ITS1) etc. herein. To guard against the possibility of PCR product contamination in PCR re-amplifications after the first, unexpected sequencing results, tools were washed, autoclaved, washed and autoclaved again before subsequent (successful) extractions. Two Agaricus-specific primers nested within the ITS1 region were designed for re-amplification: ITS1-Fwd016 (att atg ttt tct aga tgg gtt gta) and ITS1-Rev323 (gag atc cgt tgc tga aag ttg tat). All ITS PCR was carried out with an annealing temperature of 56 C (occasionally 54 C) and an extension of 1 min for 35 cycles. Sequences were obtained by the University of Pittsburgh’s Biomedical Research Support Facilities using their preferred technologies. Sequence trace files were inspected and hand corrected as needed with the Seqman module of the Lasergene version 5 package (DNAstar). Sequenced taxa were identified by alignment and comparison with sequences from known taxa. Readable portions of sequences were deposited in GenBank (EU071698–071701).
|
HISTORY |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSHISTORYRESULTS OF MOLECULAR AND...DISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
, Singer 1984). I hope to resolve several such misconceptions.
For at least 8 y a number of collectors based in or around Cambridge, Massachusetts, either corresponded relevantly with Peck or participated in the A. brunnescens saga in other ways. Two events occurred in 1892. First, in June (in W.G. Farlow’s hand, correcting "Jan" in his typed notes and handwritten specimen "tag"), Alice E. ("Mr.") Palmer provided specimens from a blacksmith’s shop in Charleston, Massachusetts, adjoining Cambridge to the northeast, to William Farlow. Farlow had the specimens drawn by Joseph Bridgham and added a note with a collection date of Feb (!) 1892. The drawing was reproduced lithographically and was incorporated into the Icones Farlowianeae (Farlow 1929) as Plate 62. The painting at FH is a bit lighter in tone and less yellow than the plate. The material originally was determined by Farlow to be A. rodmanii Peck (= A. bitorquis [Quél.] Sacc.) but later was redetermined by E.A. Burt as A. brunnescens Peck (Burt in Farlow 1929). The aspect of the specimens, captured beautifully by Bridgham, is of a member of section Chitonioides Romagn., similar to A. gennadii (Chatin & Boud.) P.D. Orton (whereas A. bisporus is a member of section Duploannulati Wasser ex Wasser). The specimens are at FH. I found this material to be distinctive, for example in the relatively low length/width (l/w) ratio of the spores (1.13–1.15 [two separate datasets, mean size 6.0 x 5.2µ, N = 60]). Farlow’s notes describe the spores as "near globose but a little longer than broad, about 5–6 µ x 6–7 µ [w x l]." In comparison typical l/w ratios for A. bisporus, A. bitorquis and A. gennadii are respectively ca. 1.18–1.28, 1.18–1.29 and 1.26–1.41(–1.69), based on multiple collections of each species. The illustrated combination of a hollow stipe and a sheathing peronate veil, virtually an appressed basal cup, is remarkable; Farlow described the stipes as solid, but to Bridgham’s eye (and to my own) they were hollow.
Also in 1892, according to George B. Fessenden in a letter to Peck dated 27 Oct 1899, Fessenden found mushrooms of "the same variety" as those that would be sent to Peck by C.G. Wells 7 y later.
On 3 Jan 1898 George E. Morris of Waltham, Massachusetts, wrote to Peck that he "hopes to send the plates of the ‘Ag. "subterraneus"’ some time this month." From the context it appears that the two men had discussed this species during a recent visit. The species name is a nomen nudum. No further information on this mushroom has been located.
In Aug (according to the box label) 1899 specimens were sent to Peck by Miss Helen M. Noyes, with her request for "the name". The information "Grows in dump heaps in Cambridge" was recorded. Peck painted the specimens (NYS 1–1349, "Agaricus brunnescens Pk" with attribution to "Auburndale, Mass/Helen A. [sic] Noyes"); as figured these mushrooms have the general aspect of A. bisporus (FIG. 2). It was the Noyes collection, according to the box label and an included note, that were designated at some point within NYS as the A. brunnescens type collection, in accordance with the protolog. It is not clear which elements of the type description are drawn from the Noyes material, for example August (per the box label) is not given as a month in which the species occurs. The published description of the annulus as "striated on the upper surface by impressions of the edges of the lamellae" is notable, as will be discussed below. This characteristic may have been present in the Noyes specimens; however Peck rendered it rather obscurely. On 2 Apr 1901 Noyes wrote again to Peck on Boston Mycological Club stationery to thank him for sending his "report for 1899."
|
On 1 Nov 1899 Hollis Webster wrote to Burt on BMC stationery referring to "Agaricus brunnescens Peck in litt. based on material from Front St. Cambridge in waste ground." On 2 Nov a postcard from "H.W." (in Webster’s hand) to Burt followed. Webster informed Burt that he was sending "a few of Ag. brunnescens Peck (in litt.) from type locality, and by same collector who sent to Peck—C.G. Wells." Webster added more information: "Flesh tends to brownish, interior soon decays, ... ring frequently double (cf. Vittadini’s figure of A. campestris)—buttons deep in soil below surface." Burt prepared his own notes on Wells’ specimens, which were added to the E.A. Burt Herbarium (now at FH). He recorded "a ring often double as in A. rodmani and with upper portion superior. —Odor of spec. rather disagreeable when received," indicating that the specimens were in poor condition; however he was able to note flesh "white with a slight ‘R06 stain and becoming more pronounced on exposure to air." Note that Vittadini figured several species under the name A. campestris; the system using the ‘R06 color designation has not been identified.
On 17 Nov 1899 Fessenden wrote to Peck in response to a letter from Peck dated 16 Nov 1899. Among other notes Fessenden states that the construction of A. brunnescens (in his and Wells’ sense) and A. maritimus (is) identical, adding that both species often mature underground. This letter appears to explain the emphasis placed on these two points in the last two paragraphs of Peck’s description.
Peck published A. brunnescens in 1900. His description is fairly complete and included spore dimensions of 6–8 x 4–6 µm.
On "7-10-1900" Morris painted specimens determined as "Ag. brunnescens". The painting, in the Peabody Museum of Salem (Morse 1918), figures mushrooms with an aspect comparable to that of Agaricus bisporus, although no hollow center is figured in the stipes of three sectioned basidiomata.
On 20 Oct 1900, after publication of A. brunnescens, Wells wrote to Peck, expressing the belief that he had provided the original specimens of the species to Peck "at suggestion of Miss H.M. Noyes (I notice you attribute them to her)." He offered to send Peck his notes from collections made in 1900 and also photos of same made by Mr Henry L. Clapp. No further correspondence, notes or photos from Wells or Clapp were found at NYS.
However at FH a description of "AGARICUS brunnescens, Pk." by Wells, dated "Boston, Nov. 10 1900" exists. It describes specimens collected 8 Oct 1900 "on rich dump, Cambridgeport, Mass." Notable was the "powdery white, striate" upper annular surface and the "lower side slightly floccose as it tears away from stem." The annulus of A. bisporus can exhibit these features although the latter is not typical. The flesh was described as "...white, changing slightly to brown at disc." The reddish brown lower stipes of both young and old specimens is unusual, but most probably reflects a rufescent reaction induced by handling. The figures referred to in this description were not located.
Fessenden later sent cultivated material of A. subrufescens and of another species to Peck (letter, 16 May 1901, NYS).
Here ends the contemporaneous story surrounding the original collections and concept(s) bearing, or potentially bearing, on A. brunnescens. Several Cambridge mycologues, including Wells, Fessenden, Webster and Burt, believed that Wells had collected and transmitted what later would become the type of A. brunnescens to Peck (the type method was specified in the American Code in 1904 and by the Cambridge Code [ICBN] in 1930: Petersen ca. 1993). These collectors had a concept of a mushroom with a strongly inrolled pileus margin, a peronate-sheathing veil or "double" annulus bearing striations (presumably imprints from the lamellae) on the upper surface and a tendency to develop beneath the substrate surface. These are generally collective characters of section Chitonioides (e.g. A. bernardii, A. gennadii) and some members of section Duploannulati (A. bitorquis and to a much lesser degree A. bisporus) and perhaps other species. The 1892 Palmer specimens at FH, which I assign to an unknown species of Chitonioides, are at least generally concordant with this concept and in Burt’s view explicitly so.
However Peck clearly attributed the crucial material to Noyes. His painting of the Noyes material is evocative of a wild A. bisporus, with an intermediate, band-like annulus having a wedge-shaped cross section. No relevant Agaricus collections or images from this period, from Wells, Fessenden or Morris have been located at NYS. Following the type method, the Noyes material would be the end of the story and the synthetic description incorporating the concepts of the Cambridge mycologists would be irrelevant if the extant type material were authentic and homogeneous. These issues will be considered below.
Twentieth century literature gradually addressed the matter of A. brunnescens. Murrill (1914) reported and illustrated a collection of "Agaricus campester hortensis Cooke" from New York. In 1922, apparently after examining the type of A. brunnescens, he wrote that "my A. campester hortensis ... seems very near this species." Stewart (1929) obtained mushrooms produced from American brick spawn purchased in New York and noted that they were "altogether different" from Psalliota (Agaricus) campestris. In the 1920s this distinction was becoming better appreciated, leading J.E. Lange (1926) to first recognize "form" bispora of Psalliota hortensis "Cooke", based on specimens collected "...on a heap of road-scrapings (soil and gravel mixed with horse droppings)." Stewart sent his cultivated material to Kauffman, whose best assessment was that it was "P. brunnescens". A.H. Smith (1940) examined the A. brunnescens type and reported spores of 6–7 x 5–6µ. His report unfortunately is notable for endorsing Burt’s and Farlow’s (1929) illustration and description as authentic. Treschow (1945) reviewed several of these reports and posed the question of "whether this [comparing A. brunnescens to the cultivars] might not be the same form." Isaacs (1967) included A. brunnescens in a list of synonyms for the cultivated Agaricus mushroom.
It was Malloch (1976) who brought the matter to wider attention, re-examining the type, reviewing much of the literature on this species and discussing other related species. Malloch concluded that "I have no doubt now that Peck’s name Agaricus brunnescens applies to what we commonly recognize as the cultivated mushroom." He further concluded that, based on priority, A. brunnescens was the correct name for the species.
Two alternate points of view (views not exhibited by "realistic mycologists", per Redhead 1992) have since appeared. First, Singer (1984) rejected synonymy of A. bisporus and A. brunnescens. Singer’s opinions appear to have coalesced around a new collection (R. Singer N5133 22 IX 1983 Evanston, Illinois; B-17943 at F) which he studied while fresh. Remarkably this mushroom, which based on my examination might belong to sec. Chitonioides, was described as having "predominatingly 2-spored basidia." Agaricus bisporus (var. bisporus) is the only species of Agaricus known to have characteristically, predominantly bisporic basidia. However several other species may develop large numbers of bisporic or trisporic basidia under certain conditions, including low temperatures and diurnal stage (Kerrigan and Ross 1987a, b; Kerrigan 1987; J.-M. Moncalvo pers comm). In fact Singer recorded "at first" a mixture of 1-, 2- and 4-spored basidia, later becoming predominantly 2-spored. This agrees with observations of Kerrigan and Ross (1987a, b) and Kerrigan (1987) on the potential for dynamic changes in basidial development and spore-number properties in Agaricus spp. over several hours. Of interest, Singer noted these changes over time, although the collection consists of only a single basidiome. Presumably the data and changes were recorded on this one mushroom after harvest because such data are challenging to develop in situ (see Kerrigan and Ross 1987a). Harvesting and storage conditions can influence spore size and basidial spore number (Kerrigan 1982); these also might have played a role with Singer’s N5133. Singer, who examined Peck’s type, explicitly took for his concept of A. brunnescens a synthesis of Peck’s and Burt’s published descriptions and Bridgham’s illustration; the latter two apply to Palmer’s Chitonioides-like material, which Singer believed to be a topotype, and to Burt’s "Cambridge view" of the species, so Singer’s determination of N5133 is not surprising. I recorded a mean spore size from lamellae in this collection of 6.2 x 5.0µ, l/w 1.24, N = 10, within the range reported by Singer.
Second, after examining the type of A. brunnescens circa 1977 I expressed reservations about unequivocally accepting the proposed synonymy between A. brunnescens and A. bisporus (Kerrigan 1987). My examination of the type emphasized fragments 1, 6, 10 and 13 (FIG. 1
). At that time fragments 4a and 4b were joined to form a single pileus 7 cm diam. At the time I assumed fragment 6, the largest stipe, to have been joined originally to the large pileus. Spores were measured from stipes and annuli, rather than lamellae, because I have found this reduces the variability of measurements. Spores deposited on the annulus of fragment 6 measured (5.6–) 6.2–6.5(–7.1) x (4.5–)4.8–4.9(–5.6) µm, mean size 6.36 x 4.87 (N = 50), l/w per each 10 spores 1.27–1.32, mean 1.31. These measurements fall at the lower limit for measurements I have taken on numerous A. bisporus specimens. I found and figured many aberrant-looking basidia and basidia that had more than two sterigmata at higher frequency than is known to be typical for A. bisporus. In fact they appeared to resemble aberrant basidia (in SEM) produced by the tetrasporic sister species A. subfloccosus (J.E. Lange) Pilát s. l. or by A. subrufescens after refrigeration at 3–4 C for several hours (Kerrigan and Ross 1987a, b). At my request Geraldine Kaye of FH obtained weather data for 1899 from Boston. Thanks to her effort I was able to write that "From October 20 to November 2, 1899, the air temperature in the Boston area fell below 3 C on four occasions." (Kerrigan 1987). This raised the possibility that the micromorphology of the type specimens might have been influenced by cold weather such that a substantial and unrepresentative proportion of bisporic basidia (and aberrant basidia) might have developed. After discussing other characters, including the relatively small spores I measured from fragment 6 of the type, I concluded, with reservations, that "the odds are better than even in favor of conspecificity" of A. bisporus and A. brunnescens, accepting Malloch’s conclusions as a "working hypothesis". Implicit was my view that not every old herbarium specimen can be determined with absolute certainty and confidence.
Malloch and I also examined the Wells 2 Nov 1899 collection in the Burt Herbarium at FH. Our measurements of the spores generally agree: 6.2–7.8 x 5.0–5.7 µm (Malloch notes at FH) vs. (5.6–)6.3–6.7(–7.6) x (4.2–)5.2(–6.1) µm, N = 20, from annulus (Kerrigan notes). These values agree with A. bisporus but the widths are too great for A. subrufescens.
Various writers and authors have emphasized variable features of the relevant collections. Color changes normally are important; however they can vary for many reasons including specimen age and condition. In some of the relevant cases the condition of the specimens was evidently poor. Therefore the degree of reddening or browning might not be a reliable means of discrimination in this case and it is difficult to evaluate the significance of recorded differences. Similarly odor, often associated with sectional-level groups, becomes useless when, as can happen with deterioration, the specimens develop an unpleasant odor. Velar development is also pleiomorphic in response to environmental and other variation. However the Cambridge emphasis on a sheathing, double or two-limbed veil in A. brunnescens, with a striate upper limb surface, for the mushroom known to Fessenden, Wells, Burt and Webster, and Bridgham and Farlow for that matter, does not compellingly invoke A. bisporus. Such an annulus is seen only rarely in A. bisporus; striations may be present on the upper surface but normally they are obscure (or fairly so) and irregular at best. In contrast in Chitonioides and in A. bitorquis this is the typical form of the veil and striations on the upper surface can be so deep and so regular as to appear to have been machined there. Thus from all available background information I had reason to suspect that different phylogenetic entities might have contributed features to a synthetic concept of A. brunnescens.
|
RESULTS OF MOLECULAR AND MICROSCOPIC ANALYSES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSHISTORYRESULTS OF MOLECULAR AND...DISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
). Fragments 4a and 4b are clearly parts of a single pileus (intact in 1977); fragment 6, a stipe, might complete this specimen; however it had somewhat smaller spores (see below). It appeared after visual reexamination that the material might be separable into two components, one of which was more fuscous, while the other had a slight yellowish cast. Based on Peck’s painting, at least three and possibly five basidiomata (minimum) were present in the original Noyes material. There is no reason to believe that the extant specimens, specifically the more fuscus elements, do not include the Noyes material. DNA amplification and sequencing.— – PCR with ITSp1 and ITSp2 or ITSp3 and ITSp4 was attempted on 12 DNA extracts from 10 of the 22 type fragments: 1, 2, 4a, 4b, 5, 10, 11, 13, 15 and 17. Only one amplification from a second extraction of the pileus fragment 10 yielded sufficient product to be sequenced directly. A total of 302 nt of ITS1 sequence was obtained; the sequence matched that of A. subrufescens Peck and included several characteristic heteromorphisms.
In other experiments fragments 10 (ITS1, 2), 13, 15 and 17 (all ITS1) yielded faint PCR product bands of typical size (and also an ITS1 band of smaller size) that after gel purification were re-amplified with the same primers along with negative controls. The re-amplified Agaricus-typical ITS1 products of fragments 10 and 17 were found to have partially readable sequences matching those of A. subrufescens. In these cases, and with the largely unreadable data from fragments 13 and 15, another ITS1 sequence was superimposed on the first. This latter sequence was also characteristic of a smaller, ~220 nt ITS1 product that co-amplified with the normal size (~350 nt) band. The anomalous sequence (=GenBank No. EU071701), obtained from the gel-purified small ITS1 product of fragment 17, had 100% identity with Bionectria ochroleuca strain CBS 102.94 (=Clonosta-Clonostachys rosea) (GenBank No. AY876924 [GenBank] .1). This result is perhaps explained by the observation that C. rosea is a common soil fungus and versatile parasite/mycoparasite that might have infected these specimens. In a follow-up experiment, amplifying first with primers ITSp1 and ITSp2, then re-amplifying with nested Agaricus-specific primers ITS1Fw016 and ITS1Rv323, and sequencing with one of the latter primers, the Clonostachys template successfully was excluded from the sequenced product and readable sequences sometimes were obtained.
A re-amplified ITS2 sequence from fragment 10 also matched that of A. subrufescens. Clean sequence reads were obtained from both strands, yielding 379 nt of data. This ITS2 sequence matched several sequences known from A. subrufescens. In contrast there were 21 single-character differences (6% not counting indels) when it was aligned with the sequence of A. bisporus RWK 1885 (GenBank No. AF432886 [GenBank] ), illustrating the ability of ITS2 sequences to differentiate the relevant species.
The assembled sequences from all reads of fragments10 and 17 are 711 and 712 nt long, due to the presence in this heterokaryon of a typical (for specimens from the Americas) length heteromorphism at about position 50 (ttt vs. tttt, see Kerrigan 2005). Following custom only the 711 nt sequence was deposited (GenBank No. EU071699). The fragment-10 assembly is effectively complete except for 20 nt between primers ITSp2 and ITSp3 in the 5.8S gene. Beginning numeration with ggaaggatca in the 18S gene through gaacttaa in the 28S gene, the fragment-17 data (EU071700) comprise positions 1–136, within the 18S-ITS1 region.
A parallel attempt was made to extract and amplify the ITS1 and ITS2 rDNA segments from fragments 1, 2, 4a, 4b, 5 and 11, most of which, based on appearance, are more likely to represent the Noyes material. No PCR products were visible on an ETBR-stained agarose gel, although positive controls worked. However the absence of a visible ITS1 product typical of C. rosea implied a different history for these fragments. When the ITS1 post-PCR reactions were cleaned and re-amplified with primers ITSp1 and ITSp2 abundant but length-heterogeneous smears of PCR products were visible on the gel. To investigate whether nested PCR could resolve a homogeneous product from the available templates, internal primers ITS1Fw016 and ITS1Rv323 were designed from the sequence of A. bisporus RWK 1885 and used on both cleaned first-round and noncleaned second-round templates. In all cases a robust band of around 300 nt was obtained. Subsequent attempts used raw first-round products (lacking visible bands on gels) as templates in a second round of PCR using the nested primers, and better sequences were obtained. These products tended to produce sequences of which only short lengths were readable.
All six fragments (1, 2, 4a, 4b, 5, 11) produced partially readable sequences that were concordant with those of A. bisporus, presenting several distinguishing polymorphisms for this species. A consensus sequence assembled from these reads (=GenBank No. EU071698) was 257 nt long, corresponding to positions 53–309 of the A. bisporus ITS1+2 sequence of RWK 1885 (=GenBank No. AF432886 [GenBank] ). The individual reads were not as clear as the best from fragments 10 and 17; consequently even in the consensus sequence there are 12 ambiguous positions and three possible errors (at positions 133, 175 and 253). Within the remaining 205 positions the A. brunnescens and A. bisporus sequences share 26 characters (~13%), even excluding indels and heteromorphisms, that are not shared with A. subrufescens. Altogether this is strong evidence for conspecificity of these fragments of the A. brunnescens type with A. bisporus.
Note that the re-amplification techniques, necessitated by the low yield of amplifiable DNA from the type fragments, are susceptible to the generation of artifacts. In most cases negative controls came up clean. However one exception was noted; this product was sequenced and found to belong to an unrelated species, whereas the A. bisporus and A. subrufescens sequences always were obtained consistently and reproducibly from specific subsets of the type fragments. For those reasons I believe that they should be considered authentic.
Micromorphology.— –
Lamellar fragments from pileus fragment 4b were examined. This material is mature and covered with shed spores but probably was no longer very actively sporogenic when dried. Cells were mostly collapsed and no cheilocystidia were observed clearly. Relatively few basidia were resolved clearly. A single sterigma was observed most often, and these frequently were hypertrophied, up to 5–6 um long. Fairly typical bisporic basidia were observed next most often, followed by a few trisporic basidia. The hymenium showed no clear instances of spore triads or tetrads, consistent with a predominance of bisporic basidia. Spores measured (5.5–)6.4–6.5(–7.2) x (4.5–)5.0–5.2(–6.0) µm, l/w = 1.27 (1.23–1.30, per sets of 10 spores), N = 30. This is near the lower end of the known range for A. bisporus (while the widths on average exceed that of almost all collections of A. subrufescens [Kerrigan 2005]). My original measurements of spores from the stipe/veil of fragment 6 were slightly smaller (see above), with means of 6.36 x 4.87 µm, N = 50. Basidia from fragment 1 included some well preserved bisporic (and fewer trisporic) basidia, plus some with less acute putative sterigmata that numbered up to four.
Fragments 10 and 13, assigned to A. subrufescens, have more spore triads and tetrads, and greater tendency toward tri- and tetrasporic basidia, and some bisporic ones and some aberrant basidia with blunt sterigmata are also present.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSHISTORYRESULTS OF MOLECULAR AND...DISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
The demonstration of fragments producing ITS1 sequences concordant with those of A. bisporus, in conjunction with morphological data from those fragments, confirms beyond reasonable doubt that a portion of the A. brunnescens type, presumably the Noyes material, is conspecific with the concept of A. bisporus. Lectotypification is made as follows:
Agaricus brunnescens Peck. 1900. Bull Torrey Bot Club 27:16 Lectotype (here designated): Fragments 4a and 4b, portions of a single pileus in the type specimen packet for A. brunnescens (NYS).
Fragments 1, 2, 5 and 11 are also concordant, based on ITS1 DNA sequences.
The presence of type fragments that yielded ITS PCR products with sequences perfectly matching those of A. subrufescens and substantially different from those of Duploannulati (in particular A. bisporus) and Chitonioides was unexpected. There are various ways in which material could find its way into an existing herbarium collection, for example by Peck’s hand. Agaricus subrufescens shares common substrates with A. bisporus: composted plant material and/or manure. The A. subrufescens fragments are fairly immature, in which stages the specimens would have been harder to identify correctly even when fresh. While the possibility of a sequencing artifact also must be considered, obtaining concordant results in independent experiments under stringent protocols and using different primers is highly unlikely. Thus I accept the A. subrufescens sequences as real data. Because my first study of the type focused largely on micromorphology of fragments 10 and 13, the former now shown and the latter suspected to be A. subrufescens, my initial observations of non-bisporic basidia therein become understandable. Fragments known and suspected to be this species now have been segregated within the boxed type.
Some historical questions remain. It is curious that the Noyes collection was made in August, when A. subrufescens is likely to fruit but A. bisporus is not. Is it possible that confusion led to the Wells material from Sep–Oct being misattributed to Noyes? Peck’s painting is not dated. What Fessenden collected, and whether any Chitonioides material ever went from Cambridge to Peck, will probably never be known.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSHISTORYRESULTS OF MOLECULAR AND...DISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
) the nether end of one such beast, which was itself a fantastical chimera. Lessons to be drawn include the need to consider nonobvious explanations for factual discrepancies and the need to be aware of the vulnerabilities of the type method. Currently, as lectotypified here, the name Agaricus brunnescens has priority over A. bisporus; however A. brunnescens itself is not necessarily the oldest available name for the cultivated button mushroom. The outcome of the proposed conservation of A. bisporus over A. brunnescens (Redhead, Kerrigan, Parra and Malloch in prep) will determine the correct binomial for the cultivated button Agaricus. It is expected that A. bisporus will be conserved because all interested parties are believed to concur. In any case Peck’s description of A. brunnescens should be treated as likely to have been augmented by inhomogeneous elements. The Palmer material at FH and its illustrations are not relevant to the concept of A. brunnescens but only to its convoluted and misunderstood history.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 E-mail: rwk{at}sylvaninc.com
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSHISTORYRESULTS OF MOLECULAR AND...DISCUSSIONCONCLUSIONLITERATURE CITED |
|---|
Edwards RL. 1990. Agaricus bisporus or Agaricus brunnescens? Enzyme Microb Technol 12:153.[CrossRef]
Falconer W. 1891. Mushrooms: how to grow them. New York: Orange Judd Co. 169 p. [1912 reprint].
Farlow WG. 1929. Icones Farlowianae. Boston: Merrymount Press. 120 p.
Isaacs BF. 1967. Studies in the genus Agaricus I. Agaricus cretacellus and its relationships. Michigan Bot 6(1):3–12.
Kerrigan RW. 1979. Studies in Agaricus I. Agaricus pattersonae. Mycologia 71:612–620.[CrossRef]
———. 1982. The genus Agaricus in coastal California (Master’s thesis). San Francisco State University. 208 p.
———. 1987. What’s in a name? The chaetaceous case of the chaste champignon. In: Wuest PJ, Royse DJ, Beelman RB, eds. Cultivating edible fungi. Amsterdam: Elsevier. p 141–154.
———. 2005. Agaricus subrufescens, a cultivated edible and medicinal mushroom, and its synonyms. Mycologia 97: 12–24.
———, Callac P, Guinberteau J, Challen MP, Parra LA. 2005. Agaricus section Xanthodermatei: a phylogenetic reconstruction with commentary on taxa. Mycologia 97: 1318–1341.
———, Ross IK. 1987a. Dynamic aspects of basidiospore number variation in Agaricus. Mycologia 79:204–215.[CrossRef]
———, ———. 1987b. Basidiospore number variation in Agaricus. In: Wuest PJ, Royse DJ, Beelman RB, eds. Cultivating edible fungi. Amsterdam: Elsevier. p 155–169.
Lange JE. 1926. Studies in the agarics of Denmark VI. Psalliota Russula. Dansk Bot Ark 4(12):1–52.
Malloch D. 1976. Agaricus brunnescens, the cultivated mushroom. Mycologia 68:910–919.[CrossRef]
Morse AP. 1918. List of the water-color drawings of fungi by George E. Morris in the Peabody Museum of Salem. Salem, Massachusetts: Peabody Museum. 70 p.
Murrill WA. 1914. Illustrations of fungi XVIII. Mycologia 6: 161–166.[CrossRef]
———. 1922. Dark-spored agarics III. Mycologia 14:200–221.[CrossRef]
Peck CH. 1884 ["1883"]. Report of the Botanist. New York State Mus Ann Rep 36:29–49.
———. 1897 ["1894"]. Report of the Botanist. New York State Mus Ann Rep 48:103–337.
———. 1900. New species of fungi. Bull Torrey Bot Club 27: 14–21.[CrossRef]
Petersen RH. ca. 1993. A brief history of the type method. http://fp.bio.utk.edu/Mycology/Nomenclature/nom-hist.htm
Redhead SA. 1992. Should we conserve this newer name? Mushroom 35(10[2]):4 [letter].
Singer R. 1984. Agaricus brunnescens Peck and Agaricus bisporus (Lange) Imbach. Mycotaxon 20:479–482.
Smith AH. 1940. Studies in the genus Agaricus. Pap Michigan Acad Sc 25:107–138.
Stewart FC. 1929. Is Psalliota brunnescens under cultivation? Mycologia 21:41–43.[CrossRef]
Treschow C. 1945. Taxonomy of the cultivated mushroom. Friesia 3:124–128.
Vittadini C. 1832. Descr. Fung. Mang. fasc. 1–3:1–64. Milano.
White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds. PCR protocols. A guide to methods and applications. San Diego: Academic Press. p 315–322.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
