[Taxacom] The Linnean algal names

[Athanasios Athanasiadis]

It is difficult to imagine the excitement or scientific boost caused  
by the advent of the Linnean collections to London. It could well be  
compared to the arrival of molecular tools in systematics, as the  
world was divided then again between those who had access to the  
collections and those, like Esper, who did not. The rumor says that  
my compatriots, who regretted their act, sent a warship to seize and  
bring back the precious cargo, but in vein. A piece of Scandinavian  
history had changed hands, but it was certainly through a straight  
affair.
	In London, the study was going on with many surprises. Amongst the  
'discoveries' made, Fucus ceranoides turned to be a fucoid (not the  
rhodophyte of Morison, Hudson, Withering, Lightfoot, Gmelin and  
others), Fucus crispus the common carrageen alga (not that of  
Hudson), and Fucus rubens a Phyllophora (still to become a Phycodrys  
in our time). The latter exemplifies the difficulties that always  
have surrounded the identity of the Linnean names, as the concept of  
typification was introduced much later and the rules changed over the  
years, while on the other hand Linne's annotations and in many cases  
the condition of his herbarium specimens left much to be wanted.
	The first authors to spend time botanizing over the algal  
collections were J.E. Smith and Dawson Turner. Smith was working on  
'English Botany' and Turner on his magnificent 'Fuci', and Linne's  
taxa were amongst the most frequently cited or discussed. Yet, both  
Smith and Turner soon realized that many of the collections were  
heterogeneous while the specimens could be neither typical nor well  
annotated. Moreover, the era of anatomical observations was still to  
come and they had neither time nor knowledge to adopt any advanced  
methods. Hence, the first errors were made as Turner did not properly  
examined the Linnean specimens - as for example his identification of  
Fucus fastigiatus with Polyides - or because he was unable to  
perceive the species diversity of certain genera, as for example in  
the case of Sargassum, where he apparently included 4 different  
species in S. natans. Other 'mistakes' were made as Turner was  
adopting Linnean concepts from later publications and not the  
protologues, as for example in the case of Ulva latissima. As  
knowledge and confusion was accumulating over the years, the ICZN  
found a simple way out, but for the ICBN priority still counts which  
of course includes the Linnean names.
	Until very recently, the identity of many Linnean names was obscure,  
when in a single publication more than 70 taxa received types (Taxon  
58: 237-360. 2009). Reading this paper, two things become immediately  
understood: the authors have overlooked the largest part of  
literature that discusses the status of these Linnean names, and  
particularly papers where purifications have been made - in some of  
these accounts the status being limited to single specimens and hence  
it should be regarded as implicit typification in accordance to  
tradition. Secondly, in most cases, there is no evidence that the  
Linnean types are in fact original material, as the relevant  
specimens lack annotations that unequivocally link to (or predate)  
the protologues. Hence such selected types are neotypes and fall  
under the regulations of Art. 9.17. Examples of these cases will be  
discussed below.

ULVA LATISSIMA. An account on the identity of U. latissima (Sp.Pl.:  
1163. 1753) was published in Taxacom (15.xii.2006) but despite this  
(and further communications) the authors (in Taxon l.c.: 255) have  
retained the erratic thesis that the name of this green alga should  
be the basionym of the brown alga Laminaria saccharina, within the  
genus Saccharina: viz. Saccharina latissima (L.) ... ex ... (in Taxon  
58: 255. 2009) - the new combination being accidentally validated, as  
on the first occasion two names were cited as basionym. The following  
must be taken into consideration: the selected 'lectotype' (a  
Laminariaceae) is annotated by Linne '4 Gotl', and while 'Gotl' is  
transliterated 'Gotland' (on the east coast of Sweden) by Savage  
(Cat.Linn.herb.1945), where no Laminariaceae has ever been reported,  
the type locality of U. latissima is Marstrand (on the west coast of  
Sweden; Linne It.Westr.: 169. 1747). Besides, the number '4' (that  
could be linked to the protologue in Sp.Pl.), is not unique as it  
also occurs on a second sheet in LINN, annotated '4// India' (Savage  
l.c.). Hence the selected 'lectotype' is not original material but a  
neotype and as such can be superseded  (Art.9.17), because it is in  
serious conflict with the protologue that unequivocally describes an  
Ulvaceae - the Laminariaceae being described elsewhere in the same  
publication and similarly detailed by Linne. Therefore, unless the  
name Saccharina latissima is conserved with a conserved type, it will  
become a later synonym of the green alga Ulva latissima - when the  
latter name will be typified with a topotype according to the  
protologue and the opinion of all authors (prior to 2008) who  
followed Linne's original account, and not 'Mant.Pl.altera' (as done  
by Gunnerus, Turner, Lyngbye or Papenfuss).

CONFERVA LITTORALIS. The case of C. littoralis (Sp.Pl.: 1165. 1753)  
is more complicated but happily does not involve mistakes of similar  
taxonomic magnitude. Here the authors (in Taxon l.c.: 244) state that  
several workers discussed the taxonomy without selecting a type,  
crediting Kylin's opinion (in Blumea suppl.II: 70. 1942) to me, and  
disregarding the fact that Linne (Fl.Svec.2 ed.: 436. 1755) limited  
the concept of this species to two elements of which one is an  
illustration, viz.:

'1170. CONFERVA (littoralis) filamentis aequalibus ramosissimis  
elongatis scabriusculis. “Spec.plant.1165.”	.
		[Conferva maritima filiformis longissima fusca, ramis alternis  
simplicibus brevioribus pallidis. Fl.Suec.1028. It.goth.261.n.3.]
		[Conferva marina capillacea longa ramosissima. Dill.musc.23.t.4.f.19.]
		“Conferva marina tomentosa (minus tenera) & ferruginea. Dill.musc. 
19.t.3.f.13.” Fl.Suec.1028.
		“Gotlandis Ylle.” Habitat “frequens in mari“ [in Europae marinis  
rupibus.]'

[Deletions from the protologue between brackets, additions between “ ”].

The above account should be regarded as an implicit typification,  
because it includes a single illustration (i.e. Dill.musc.19.t.3.f. 
13) based on material from England, and a diagnosis based on material  
from Gotland (in the Baltic). The Dillenian element was also included  
in the protologue, being cited as 'Fl.Suec.1028' after the Linnean  
polynomial 'Conferva maritima filiformis longissima...' (that  
actually refers to 'Fl.Suec.1029').
Kylin (l.c.) realized the error, but disregarded Linne's (1755)  
purification, concluding that:

'Die von Linne [1753] beschriebene Conferva littoralis enthält zwei  
verschiedene Komponenten: erstens eine Ostseekomponente [i.e. Fl.Suec. 
1029] und zweitens eine Komponente aus England [i.e. Dill.musc.23.t. 
4.f.19.] ...die Ostseekomponente Dictyosiphon foeniculaceus ist und  
die Komponente aus England (= Dillenius Taf.4, Fig.19) eine  
Cladophora-Art darstellt.'. [my annotations in brackets]

While it can be questioned whether no1028 or no1029 aimed to be in  
the protologue, both were cited - the former as 'Fl.Suec.1028' and  
the latter as the polynomial itself, together with 'Dill.musc.23.t. 
4.f.19'. Of the two Dillenian syntypes, Linne (1755) retained the  
former (Dill.musc.19.t.3.f.13), which should be the apparent  
lectotype. The identity of this illustration is not clear as both  
Pylaiella and Spongomorpha have been suggested to be, but most  
important is the fact that Dillenius specified his material to grow  
at the base of rocks (as his illustration suggests too) and hence if  
Pylaiella (which is the logical choice) it must be associated with  
Pylaiella rupincola (Areschoug) Kylin (= P. littoralis) rather than  
P. kylinii Du Rietz - the latter species growing on Ascophyllum or  
Fucus and exhibiting an alternation of generations (according to  
several workers, such as Knight, Kylin, and Nygren). In contrast, the  
selected lectotype (in Taxon l.c.: 244) is the Dillenius illustration  
in plate 4, fig. 19, and despite the fact that even this element was  
specified (by Dillenius) to grow on rock, as epitype was selected (in  
Taxon l.c.) material growing on Fucus. Accepting Linne's (1755)  
purification as a typification (a case to be considered at the next  
Botanical Congress), a new epitype can be selected (and this time  
hopefully in agreement with Dillenius). If not it remains to be shown  
that the present epitype represents the same entity that grows even  
epilithically.

FUCUS INFLATUS. The undated and lacking locality 'lectotype' of Fucus  
inflatus is considered by the authors (in Taxon l.c.: 249) to be a  
synonym of Fucus vesiculosus (following Powell's opinion from 1957),  
despite that a later study of 'morphometric characters indicate...en  
equal probability of being F. [inflatus]' (J.mar.biol.ass.U.K. 65:  
449. 1985). The latter actualizes Foslie's thesis (in  
Troms.Mus.Aarsheft.9: 109-111. 1886) that it is highly unlikely that  
Linne overlooked this conspicuous element of the Lapland flora,  
describing instead two forms of F. vesiculosus as two separate  
species. Besides, both the epithet inflatus and the diagnosis itself  
fit best the relevant species rather than F. vesiculosus (Foslie  
l.c.). Because the status of the 'lectotype' as original material is  
questionable, it should be superseded by a topotype that fits best  
the protologue - if a modern analysis would still indicate that the  
specimen in LINN annotated 'inflatus 5' (Savage l.c.) is indeed F.  
vesiculosus.

While the above examples show that we are still far from establishing  
or securing the identity of several Linnean names, it is surprising  
to read that our colleagues suggest possible rejection of those names  
whose identity may 'threaten' existing ones. The obvious question  
here is why the Linnean names, when there are so many other of  
Hudson, Withering, Stackhouse, or Roth, that similarly 'threaten',  
and is it not conservation the procedure we have adopted over the  
years, when a 'threat' becomes real which allows the interpretation  
of all information on a new basis ?

A. Athanasiadis,
Goteborg, Sweden