A NJMA Foray collection (SFF at Kittle Field). A single collection/fruitbody found by an unknown foray participant in a mixed hemlock-pine-hardwoods forest.
L. variicolor is a possibility, too, as it also occurs in the USA (see obs 219692 and obs 256646). For a bona fide collection of snellii identified by Dr. R.E. Halling, see obs 219324.

Spore Measurements & Discussion:
> Not from a spore print, but extracted from the tubes of the exsiccata with 5% KOH (aq), and measured only in the profile (dorsi-ventral) view.
[21/1/1]; L x W = (14.4-) 16.4~20.9 (-22.3) x (5.1-) 5.6~6.7 (-7.0) μm, L’ x W’ = 18.4 × 6.1 μm; Q = (2.70-) 2.73~3.38 (-3.46), Q’ = 3.03.
Smooth, dingy yellow to brownish-yellow, with one very large and irregularly shaped oil drop or several overlapping large ones; mostly subfusiform, spindle-shaped, sometimes moderately ventricose, sometimes symmetrically constricted in the area of the suprahilar plage in the face (dorsal/ventral) view; somewhat inequilateral, with a fairly shallow suprahilar plage in the profile view.
> Smith & Thiers report (15) 16-22 × 5.5-7.5 μm for L. snellii.1 For L. variicolor, den Bakker & Noordeloos give (10.0-) 13.5 -17.5 (-20.0) x 5.0-6.5 pm, Q = 2.4-3.1, Qav = 2.6-2.8 (-2.9).2 It’s clear that the spore dimensions of MO288473 are in much better agreement with those of L. snellii.

DNA Sequencing Discussion and Results:
> Originally nrITS was requested for this suspected L. snellii collection in order to compare it with the ITS sequence of the genetically confirmed Leccinum variicolor voucher, obs 256646. However, most frustratingly, the ITS reads of MO288473 suffered from hopeless heterogeneity, so its nrLSU sequence was procured instead.
> A clean and contiguous nrLSU sequence consisting of the first 967 bps (through to and including the LR5 primer) was obtained from this collection and uploaded to this post. There are no ambiguous characters, so it’s a single haplotype.
> A BLASTn search of this LSU sequence specifically targeting Leccinum snellii returned only two tiny sequence fragments (<10% query cover) that are totally useless for comparison.
> A BLASTn search of this LSU sequence specifically targeting Leccinum variicolor returned only two useful accessions – AF139706 and KF112445 – with query cover of 89% and 85%, respectively, and 97.6% and 98.1% similarity to MO288473, respectively. Even if at least one of these vouchers were properly identified, then MO288473 is unlikely to be L. variicolor based on the above degree of genetic divergence.
> Finally, a general BLASTn search of this LSU sequence returned a list with 44 Leccinum accessions as top hits. A phylogram of these hits was generated in GenBank and posted to this observation. Boletus roseoflavus and Tylopilus felleus serve as an out-group. When the list is sorted by % identity, the top 4 hits are >99.4% similar to MO288473.

> According to the pioneering research of den Bakker and Noordeloos on the Leccinum of Europe, L. snellii (an exclusively North American taxon) and L. variicolor (a European taxon that also occurs in North America) belong in subsection Scabra, a phylogenetically delimited group of mostly brown-capped taxa associated strictly with birch trees.2 Their Gapdh and ITS2 sequences are so different as to suggest the two taxa are not even closely related despite their remarkable macro-morphological likeness.2 In the Gapdh and ITS2 phylograms of subsection Scabra, L. variicolor is shown to split off from the rest of the tree as a distinct satellite/basal branch. L. snellii, however, nests deeply in the main tree; it is closely related to L. schistophilum in Gapdh and the L. scarbrum/melaeum/rotundifoliae cluster in ITS2. Without the use of DNA, the two taxa can be separated based on the microscopic features of the pileipellis and the ecology/habitat.2 Interestingly, den Bakker and Noordeloos make no mention of the fact that the spores of L. snellii are on average noticeably longer and wider that those of L. variicolor (vide supra).
> Though LSU is not the best genetic marker for generation of single-locus phylogenies of a single genus, for it possesses a barcode gap inferior to that of ITS and protein-coding genes to separate closely-related taxa,3 surprisingly, the GenBank cladogram doesn’t look all that bad. Surely, there are weird inconsistencies, probably related first and foremost to incorrect identification of some vouchers, poorly edited sequences and BLAST algorithm “quirks”, but therein one can also see some real and recognizable clading patterns corresponding to the generic subsections of Leccinum delineated by den Bakker and Noordeloos.2 For example, subsection Scabra, seen at the very top, is represented by vouchers scabrum, rotundifoliae, and schistophilum, while subsection Leccinum (i.e., orange/red-capped taxa with marginal flaps of sterile tissue) is the sister-group directly below, featured by its bona fide members aurantiacum, cerinum, manzanitae, quercinum, and vulpinum; finally, Leccinum duriusculum is the sole representative of subsection Fumosa.
Thus, based on the above discussion, MO288473 was also expected to clade with other representatives of subsection Scabra in the LSU cladogram, but instead it nests within a residual assemblage of accessions not associated with any recognized subdivisions of Leccinum Gray and consisting of vouchers of unidentified taxa, as well as those of dubious identification.
Interestingly, the two GenBank accessions of purported L. variicolor are neither part of the subsection Scabra group in the phylogram, nor are they clading together – both an indication that they may not be conspecific. Indeed, a separate alignment of AF139706 and KF112445 shows the two sequences to be very different. It remains to be seen if either of these sequences is identical to that of Leccinum variicolor voucher, MO256646.
Lastly, here are a few words about the top 4 hits. Two of them are vouchers of L. flavostipitatum, one is labeled as L. scabrum (likely a misidentification), and the remaining one doesn’t even have a name. Regarding L. flavostipitatum – both as a formally published species and as the proposed identification attached to the top BLASTn hit (AF139696) – Michael Kuo, who has had a lot of interest in North American leccinoid fungi and discussed their current state of taxonomy on his website, writes:
“Aside from the type collection(s), only one collection labeled Leccinum flavostipitatum is found in online herbaria (in NY, from Maine, in beech-hemlock woods). But despite being poorly documented Leccinum flavostipitatum is potentially very important since it appears to be a conifer-associated species that lacks overhanging flaps – which, if true, would place it subsection Scabra or Fumosa of section Leccinum as the only known (so far) conifer-loving species in either subsection. In separate studies by den Bakker & Noordeloos (2005) and Binder & Hibbett (2004), the DNA of a specimen identified as Leccinum flavostipitatum and deposited in GenBank by Binder & Besl (AF139696, with no voucher citation or collection location) grouped in section Leccinum, fairly clearly separated from subsection Leccinum but not definitively placed in either subsection Scabra or Fumosa. Misidentification of the specimen should not be ruled out, but the preliminary result definitely warrants further investigation.”4

Conclusion: Macro-morphologically, ecologically, and micro-scopically MO288473 is likely to be L. snellii. Unfortunately, molecular characterization of this collection has not yet confirmed its identity as such because of the heterogeneous nrITS sequence and lack of nrLSU sequences of bona fide L. snellii in GenBank for comparison.

1 A.H. Smith and H.D. Thiers, “The Boletes of Michigan”, p. 208.
2 H. den Bakker and M. Noordeloos, “A Revision of European Species of Leccinum Gray and Notes on Extralimital Species”: Persoonia 2005, 18(4), pp. 511-587.
3 C.L. Schoch et al., “Nuclear Ribosomal Internal Transcribed Spacer (ITS) Region as a Universal DNA Barcode Marker for Fungi”: PNAS 2012, 109 , pp. 6241-6246.

Species Lists


T0 (slicing began) + 15 seconds
T0 + 32 seconds
T0 + 127 seconds
T0 + 339 seconds
A GenBank-generated BLASTn phylogram of associated nrLSU hits
Mounted in KOH and viewed at x1000; 1 div. = 0.465 μm; the longest observed spore of 22.3 μm is shown
Mounted in KOH and viewed at x1000
Mounted in KOH and viewed at x1000

Proposed Names

76% (2)
Recognized by sight: An NA taxon that looks very similar to the European L. variicolor

Please login to propose your own names and vote on existing names.

= Observer’s choice
= Current consensus


Add Comment
Observation Notes updated (25-Jan-18)
By: I. G. Safonov (IGSafonov)
2018-01-25 19:46:08 CST (-0500)

Added spore measurements & pictures and DNA sequencing results & discussion.

Thanks, Dave
By: I. G. Safonov (IGSafonov)
2017-08-29 11:43:04 CDT (-0400)

In this one, the red staining in the cap took less than 1 min to begin developing. Maybe we should rename snellii to patrioticum. :-)

Nice sequence, Igor.
By: Dave W (Dave W)
2017-08-29 07:45:33 CDT (-0400)

The staining on the context of snellii can take awhile to develop, especially the red in the upper portions.