The identity of Tachina westermanni Wiedemann, 1819 (Diptera: Calliphoridae or Tachinidae) with a solution to a nomenclatural problem.

Tachina westermanni Wiedemann, 1819 was based on four syntypes, two conspecific calliphorids and two conspecific tachinids. Two existing but contradictory lectotype fixations have resulted in confusion as to the correct application of the specific name westermanni Wiedemann. Evidence is presented showing that the lectotype fixation of Townsend in 1931 is valid and assigns westermanni Wiedemann to the Calliphoridae, with Pericallimyia westermanni as the valid binomen. The valid name for the tachinid taxon becomes Brachelia westermanni Robineau-Desvoidy, 1830 and a neotype is designated for it in the interests of nomenclatural stability.


Introduction
In the Catalogue of the Diptera of the Afrotropical Region (Crosskey 1980a) there are two taxonomic species based on the same nominal species, Tachina westermanni Wiedemann, 1819 from South Africa. The first is listed by Pont (1980: 795) in the Calliphoridae as Pericallimyia westermanni (Wiedemann) and the second is listed by Crosskey (1980b: 846) in the Tachinidae as Brachelia westermanni (Wiedemann).
Tachina westermanni was described by Wiedemann (1819: 23) on the basis of an unstated number of males ("♂. Prom. bon. sp. [Promontorium Bonae Spei = Cape of Good Hope] Januario"). Subsequently it was included in the second part of his Aussereuropäische zweiflügelige Insekten (Wiedemann 1830: 291). Again, he gave no indication of the number of specimens he had before him, but this time stated that the material he examined was "In Westermann's und meiner Sammlung.". As will be explained below, Wiedemann had at least four specimens before him, thus all syntypes. Zimsen (1954: 21) reported the existence of three specimens of Tachina westermanni in Westermann's collection in ZMUC. In a footnote to the entry that is marked with an asterisk on p. 21, Zimsen (1954: 8) noted that one of the specimens was in a separate case from the two others.
It has been known for a long time that the syntypic series of T. westermanni is mixed and consists of two different species. The material in Copenhagen consists of two conspecific calliphorid specimens and one tachinid specimen, whereas the material in Vienna consists of a single tachinid specimen, conspecific with the one in ZMUC. Two authors, Townsend (1931b) and Emden (1960), have designated lectotypes, explicitly or implicitly, for T. westermanni, with the result that the name applies to either the calliphorid or tachinid species depending on which lectotype designation is accepted as valid. This has led to the current situation in which the name is treated as valid for both species, in different families, a fact that went unnoticed for 35 years. The purpose of the present paper is to discuss the merits of the actions of Townsend (1931b) and Emden (1960) and argue that Townsend proposed the valid designation. In addition, we propose a neotype for the nominal taxon Brachelia westermanni Robineau-Desvoidy (not Wiedemann) to preserve the name Brachelia westermanni for the tachinid taxon involved and ensure nomenclatural stability.
Material. One of us (KR) examined the three syntypes in ZMUC. The syntype in Vienna was examined by JEOH during a visit to NMW in 2012 (O'Hara 2013) and by all of us from photographs of the specimen and its labels taken by JEOH. JEOH also photographed the holotype of Loewia sycophanta Schiner (NMW). A male of Pericallimyia westermanni (Wiedemann) recently captured by PC in South Africa (MZUR) was studied by KR.
Format for citing labels. Specimen labels are numbered successively from the highest to the lowest on the pin, the numbers being enclosed within parentheses. The lines on each label are separated by a slash (/). If the label text itself contains a printed or a handwritten slash, then all the label lines are separated by double slashes (//). The label text is usually cited without use of quotation marks. For one specimen the labels are distributed on three pins. Our comments are given in square brackets. For some labels both comments and label translation are given in square brackets, in such cases we have indicated explicitly what is a translation and what are our comments.
Photographic methods. The techniques used for preparation of Figs. 1-6, 9-11 were explained in Rognes (2012Rognes ( , 2013. The images in Figs. 12-19 were taken with a Sony NEX-C3 digital camera equipped with a 30mm macro lens and arranged into plates using Adobe Photoshop CS5. The method used by ZMUC for making the photographs used in Figs. 7 and 8 is not known.
Wiedemann (1830: 291) mostly repeated his 1819 description but used the word coal-black ("kohlschwarz") when describing the mesonotum and the pleura, and wrote that the frons ("Stirn") was black, not using the term "Stirnmitte" employed in 1819. On p. 280 he stated that "… [e]s bleibt daher kein beständigeres Merkmal für die Gattung Tachina, als die stets nackte Fühlerborste [there does not exist a more reliable characteristic for the genus Tachina than the bare arista]", implying that a bare arista also characterizes his Tachina westermanni. He also added that the species was "Vom Kap [from Cape]" and was present in Westermann's and his own collection ("In Westermann's und meiner Sammlung"). This mention of his own collection was not given in the 1819 paper. Wiedemann's own collection was subsequently acquired by NMW (Pont 1995 . Condition: the specimen is in excellent condition, but the abdomen has a purplish hue (due to its age, almost 200 years) not found in the recent specimen in MZUR. The pin is long and slender.
Syntype 2 (Calliphoridae, ZMUC, Fig. 6). Labels: (1) ♀ [printed on small square label]; (2) Mus. / Westerm. [printed]. Condition: the specimen is in excellent condition but lacks the left mid tibia and tarsus, and the abdomen has a purplish hue as noted for syntype 1. It lacks a determination label but the big Westermann label under syntype 1 (Fig. 5) is clearly meant also for syntype 2. The pin is similar to the one on syntype 1.
Syntype 3 (Tachinidae, ZMUC, Figs 7-11). Labels: (1) Wiedemann's hand] [labels 2-4 on a separate pin beside specimen, there being no space for more labels on the pin carrying the specimen]. Condition: this syntype is not in the same good condition as the other Westermann specimens. The integument of the scutum is partly destroyed as shown in the ZMUC photograph (Fig. 7). The pin is very short and with a peculiar pin head (Fig. 7), suggesting it might have been collected on another occasion than syntypes 1 and 2. The head of the syntype was glued to a card beside the specimen when received by KR, whereas it was attached to the body in the photographs from ZMUC (Figs. 7,8 . Condition: to judge from the photographs, the condition of the syntype is fair but the first flagellomeres are missing (Fig. 14).

The syntypic status of the four specimens
It is our contention that the aforementioned four specimens were seen by Wiedemann when he described Tachina westermanni and therefore have the status of syntypes. This is in agreement with the interpretations of Zimsen (1954), who examined the three specimens in ZMUC, and Emden (1960), who examined all four specimens. Below we discuss this matter in some detail.
Syntype 1 carries a big label including the text "[…] Jan. 1817.". This label very likely also applies to syntype 2, both also carrying a similar "Mus. / Westerm." label. Neither syntypes 3 or 4 have a date on their labels. It follows that Wiedemann saw the two calliphorid syntypes when he wrote his description, which included the date "Januario".
Syntype 3, which is the specimen Zimsen (1954: 8) stated was kept in another case than the two others, carries a determination label written by Wiedemann himself (Fig. 11), so he presumably had this specimen before him. Syntype 4, even though bearing only one label possibly written by Wiedemann (Fig. 15, fourth from top), was once part of his own collection and hence was likely seen by him as well. Syntype 3 was accepted as a syntype by Zimsen (1954), who evaluated it within the context of a broader study of the insect types of Wiedemann in ZMUC. Both syntypes 3 and 4 were accepted as syntypes by Emden (1960), who carefully considered their status during his study of the identity of Tachina westermanni.
In light of the conclusions of these earlier assessments, and in the absence of evidence to the contrary, we accept both tachinid specimens as syntypes. All the syntypes are dark bluish, with a thin layer of white microtrichosity shifting with angle of view, and with a red-tipped abdomen (Figs. 4, 8, 13). Thus, all four syntypes satisfy this part of the original description. The abdomen of the first calliphorid syntype is now rather purplish on the sides, as shown in Fig. 2 (possibly due to its age). On the other hand, other parts of the description such as the reddish palpus, the bare arista, and the first flagellomere not being longer than the pedicel, only apply to syntypes 3 and 4, as noted by Emden (1960). The palpus of the calliphorid syntypes is currently somewhat reddish (after almost 200 years), but in the recent specimen in MZUR the palpus is very dark with at most the extreme tip slightly lightened. In syntypes 1 and 2 the first flagellomere is about 3x the length of the pedicel and the arista is long plumose (Fig. 3). Emden (1960) also suggested that Wiedemann's description "Stirnmitte schwarz" (1819) and "Stirn schwarz" (1830) meant that the frons was of considerable width, which he considered not to be the case for the calliphorid syntypes. It is correct that in the calliphorid syntypes the eyes are more closely approximated than in the tachinid syntypes (compare Figs. 3 and 9), but even though the frontal vitta of the old calliphorid syntypes is now rather dark brown (Fig. 3), there is a black frontal vitta in the recent specimen examined. We assume it was black in the calliphorid syntypes at Wiedemann's time, and that Wiedemann's words regarding the frons colour were applicable also to the calliphorid syntypes.
Townsend's treatment of Tachina westermanni Wiedemann, 1819 Townsend spent six months in 1928 visiting European and American museums for the purpose of studying original type specimens of "oestromuscoid" flies (Townsend 1931a: 157). The results of the visits were presented in two "notes", the first embracing the New World types (Townsend 1931a), the second the Old World types (1931b, 1932). From his visit to Copenhagen he reported on Tachina westermanni Wiedemann as follows (Townsend 1931b: 375): This statement, without any doubt, refers to the two specimens listed above as syntypes 1 and 2 and housed in the Westermann collection. Not only does Townsend cite the original label on syntype 1 correctly (including the date) (Fig. 5), but he also notes that the "male Pt" is wrongly identified as a female. This latter statement refers, also without any doubt, to syntype 2, which carries a label with a printed "♀" (Fig. 6).
The simple fact that the two calliphorid males are distinguished by different sex labels and that Townsend (1931b) identified the "male Pt" as the one with a female label is enough in our view to fix the other specimen, Townsend's "Male Ht" (which even carries the label text cited by Townsend), as a lectotype. This specimen is readily identifiable as Townsend's "Male Ht" (the lectotype) even though it does not carry any labels by Townsend to this effect. Townsend's action is a lectotype fixation, i.e., an implicit lectotypification, according to the terminology of O' Hara et al. (2009: 10).
For lectotypes designated before 2000, the Code Article 74.5 (ICZN 1999) states: In a lectotype designation made before 2000, either the term "lectotype", or an exact translation or equivalent expression (e.g. "the type"), must have been used or the author must have unambiguously selected a particular syntype to act as the unique name-bearing type of the taxon. When the original work reveals that the taxon had been based on more than one specimen, a subsequent use of the term "holotype" does not constitute a valid lectotype designation unless the author, when wrongly using that term, explicitly indicated that he or she was selecting from the type series that particular specimen to serve as the name-bearing type.
It is our opinion that Townsend (1931b: 375) "unambiguously selected a particular syntype to act as the unique name-bearing type of the taxon" in question, i.e., Tachina westermanni Wiedemann, 1819. The statement "new genus published elsewhere" by Townsend (1931b: 375) refers to the nominal genus Africomusca published by Townsend (1933: 441 (Pont 1980). The type species of Africomusca ("Genotype" in Townsend's phraseology) is cited as "Tachina westermanni Wied-Cape Good Hope", and the types of this species are cited as the "Male holotype and paratype, Copenhagen (Westermann Coll.) […]", the very specimens listed by Townsend (1931b: 375) from Westermann's collection in ZMUC. Townsend (1933: 441) Zimsen (1954: 8) emphasized that Townsend saw only these two calliphorid syntypes and did not see syntype 3 in ZMUC (a tachinid). This does not affect Townsend's fixation of a calliphorid syntype as lectotype because the Code requires only that the lectotype be an original syntype (cf. Article 74.5's mention of "syntype" and "the type series"); there is no provision that the designator must have seen all existing syntypes.
We thus consider syntype 1 in ZMUC as the lectotype of Tachina westermanni Wiedemann, 1819 and treat this name as valid in the combination Pericallimyia westermanni (Wiedemann, 1819), as used by Zumpt (1956: 83), Pont (1980: 795) and Sabrosky (1999: 126 The "Male Ht and male Pt" seen by Townsend (1931b) have been given (by KR) red labels as lectotype and paralectotype, respectively. Syntype 3 has also been given (by KR) a red paralectotype label. The fourth syntype (in NMW) (also a tachinid) has not been given such a label by us.
To Emden (1960: 403) it was "perfectly obvious" that most parts of Wiedemann's descriptions in 1819 and 1830 were not based on "the species to which Townsend assigned the name westermanni Wied."; i.e., the calliphorid Pericallimya westermanni (Townsend). Making a tabular comparison between Wiedemann's and Townsend's descriptions, he concluded that Wiedemann's description best fitted the tachinid syntype in ZMUC (syntype 3), which was correctly interpreted as conspecific with syntype 4 in NMW and the holotype of Pseudoloewia sycophanta, also in NMW. Emden (1960) went on to designate the tachinid specimen in ZMUC (Figs. 7-11), "which is labelled in Wiedemann's handwriting 'Westermanni Wied Cap. bon. sp. Westermann' [not entirely correctly cited, cf. Fig. 11] as the type of Brachelia westermanni in agreement with Robineau-Desvoidy's and Brauer and Bergenstamm's identifications". Emden here explicitly designated syntype 3 as lectotype of Tachina westermanni Wiedemann.
Emden discussed but did not recognize Townsend's implicit lectotype fixation for two reasons. First, he objected that "in this paper Townsend [1931b] only lists a 'male Ht and male Pt' as present in the Westermann Collection, without stating in this paper (or any other paper belonging to that series) that he proposes to designate types" [italics in the original] (Emden 1960: 404). This objection is not valid. Numerous implicit lectotypifications of this sort have been accepted (e.g., Crosskey 1969: 88, O'Hara et al. 2009 and they are valid, provided the author gave additional information, besides the mere mention of "holotype", that makes it possible to identify the intended specimen in collections without any doubt.
Emden's second objection follows directly after his first: "Nor could a wholesale selection of types without comparison with the descriptions be acceptable and valid.". This objection is allied to the criticism made earlier in the paper, to the effect that Townsend, "in the face of the important discrepancies between Wiedemann's descriptions and the specimens seen by himself […] was not justified in declaring one of them to be the holotype in contradiction to the typical specimen in the Wiedemann-Einthem [sic] collection on which Brauer and Bergenstamm based their statements". This objection is also not valid. Designating a lectotype is a nomenclatural and not a taxonomic act. The Code requires that the lectotype must be an original syntype, but there is no provision that it must be the specimen that most closely matches the description. In this case Wiedemann's descriptions undoubtedly match Townsend's lectotype in some, but not all, respects.
We therefore reject Emden's objections to Townsend's lectotype fixation and consider it as valid. This has the consequence of invalidating Emden's own, and subsequent, lectotype designation. Consequently, the use by Crosskey (1980b: 846) of Wiedemann's name westermanni in the combination Brachelia westermanni for the tachinid which has Loewia sycophanta Schiner as a synonym is therefore also not valid. This means that the tachinid "Brachelia westermanni" of Crosskey (1980b)
[The antennae do not reach the lower facial margin; the two last segments equal in length: the third cylindrical, not thicker than the second. First segments of arista very short. Oral margin quite long, lower facial margin projecting; oral margin with setae; body cylindrical, with bluish quite shining colour.] This is a tachinid, not a calliphorid, genus. The type species by monotypy, "BRACHELIA WESTERMANNI. R.D." (Robineau-Desvoidy 1830: 62), was described both in French and Latin as follows (citing only the French text): Note the description of the abdominal tip ("anus") as red. Robineau-Desvoidy saw only a single specimen of his new nominal species: "Le seul individu que je connaisse provient du cap de Bonne-Espérance, et fait partie de la collection de M. Dejean. M. Latreille l'avait étiqueté Tachina Westermanni".
The Diptera part of Dejean's collection went to Bigot, and "what remains is in OUMNH" (Evenhuis et al. 2010: 238). According to Pont (personal communication to KR) the Tachinidae part of the Dejean collection was given to BMNH. The holotype of B. westermanni has not been found in OUMNH (Pont, personal communication to KR) or in BMNH (Wyatt, personal communication to KR) and is assumed lost. But it is fairly certain, at least not unreasonable to assume, that the holotype of this nominal species belonged to the same taxon as the two tachinid syntypes of Tachina westermanni Wiedemann and the holotype of Loewia sycophanta Schiner (as treated by Crosskey 1980b: 846).
When Tachina westermanni Wiedemann was (erroneously) considered a tachinid taxon belonging in the genus Brachelia Robineau-Desvoidy, the name Brachelia westermanni Robineau-Desvoidy became a junior secondary homonym and junior subjective synonym of Tachina westermanni Wiedemann. Now that Wiedemann's name Tachina westermanni applies to a calliphorid species, Brachelia westermanni Robineau-Desvoidy becomes the valid name for the tachinid species. The only necessary action is to fix the identity of this name by designating a neotype to replace the lost holotype. In our view, the best specimen to select as neotype is the specimen in ZMUC that Emden (1960: 404) invalidly designated as lectotype for Tachina westermanni Wiedemann. This will preserve the use of the name Brachelia westermanni for the tachinid, even though the specific epithet is now attributed to Robineau-Desvoidy and not Wiedemann.
We hereby designate as neotype for Brachelia westermanni Robineau-Desvoidy a male in ZMUC (Westermann's collection) (Figs. 7-11), with these labels: (1) Wiedemann's hand] [labels 2-4 on a separate pin beside specimen, there being no space for more labels on the pin carrying the specimen]. One of us (KR) has added a red neotype label on a third pin beside the specimen. This pin thus carries two red labels: one as paralectotype for Tachina westermanni Wiedemann, and a second as neotype for Brachelia westermanni Robineau-Desvoidy. The neotype label is on top.
Type locality: "Cap bon Sp." [Cape of Good Hope], the same as the lost holotype.