Appendix 6 – A lucky evening, 12 October 2000, when the Colias ponteni pupa was found!; Appendix 7 – Bengt-Åke Bengtsson's clarification of how the 11 known specimens should be named

A fortunate evening on 12 October in the year 2000 when more Colias ponteni were found!

Göran Sjöberg has just discovered an additional 2, previously unknown, specimens of Colias ponteni and moreover a pupa, probably also of Colias ponteni, at the Evolution Museum in Uppsala following the museum's extensive renovation. Göran surrounded by his friends Lars Hedström, Clas Källander, Stefan Eriksson and Ingemar Frycklund. Photo: Jan Isidorsson.

# Appendix 7

Bengt-Åke Bengtsson's clarification regarding how the 11 known specimens of the butterfly Colias ponteni should be designated

I have reread Wallgren's description of Colias ponteni and see that he does not specifically mention the number of specimens at his disposal. But based on what you write about the placement of the type material and other matters, I assume that it was precisely three specimens Wallgren had before him. Let us proceed from that.

In that case, only these three specimens can be included in the type material of ponteni and be designated syntypes – until a lectotype and two paralectotypes are designated and published in some "recognized" journal or in a book. I would not include further specimens, found subsequently, since one cannot count on Wallgren knowing of them (according to your findings). Well, as additional material further specimens could very well be included. This would then apply to the two specimens that Samuel Pontén himself retained for his brother's account, which are now at the Evolution Museum in Uppsala. If the types should disappear for any reason, one would still have additional material to designate a neotype for the species. It should be chosen with care, that is, preferably from material derived from the same area and preferably from (approximately) the same time as the original specimens come from. In the future article in which this is published, the account of the pupa can naturally be included. (….)

Regardless of whether the specimens at BMNH derive from material collected by Pontén in 1852 or not, they cannot be regarded as syntypes of ponteni, unless someone has already designated lectotype and paralectotypes and published this. When an article on ponteni is published, imperialis should naturally be mentioned as a junior synonym. One need not mention more than clarifying the arguments that show that it is indeed the same species. One can of course also present the known and unknown circumstances surrounding the discovery of imperialis. These specimens should not be included in additional material, since they have already been described under another name.

Regarding the three specimens of Colias ponteni deriving from the Felder and Elwes collections now at BMNH, the following applies: They are still not "syntypes" either of ponteni or of imperialis, but are merely specimens that should be assessed to be one or the other species. If it could be proved that these three specimens come from the same locality (in the narrow sense) as the type series, that is, Wallgren's 3 described Colias ponteni, they could possibly be called topotypes. Topotypes are specimens subsequently collected at the type locality.

Nowadays one customarily designates a suitable specimen as holotype, and the remainder become paratypes (regardless of sex). Previously one used the concept allotype, but that has been abandoned. I am not certain whether there is an exact year when the concept holotype began to be used. I would guess it came into use in strict form first in 1961, when the first version of the International Code of Zoological Nomenclature (ICZN) was published. Before that one simply used the concept type. If a single specimen was presented and described but not labelled with any type label, that specimen will become a holotype. If, on the other hand, there were several specimens (syntypes) in the material that formed the basis for the description, one must designate a lectotype and select paralectotypes from the remaining material.

In principle, a species can be newly described on the basis of any stage. But in order for other researchers to be able to compare their material with the type material, one should avoid having in one's redescription a holotype that is difficult or impossible to identify. Therefore one should preferably use fully formed animals (in the case of insects) which are usually the easiest to determine.

Every genus that is described must have a type species. The type species for Colias is hyale. The type species for the subgenus Protocolias is imperialis if the claim in Dábreras's book is correct that imperialis was designated by Petersen in 1964 as the type species for this subgenus. See however below on this!

There have always been "splitters" and "lumpers". Every time someone is to produce a dissertation on a group of butterflies, it seems as if the person in question must split up a known genus into (multitudes of) new genera. A famous (infamous!) example is the Russian lepidopterists' subdivision of the case-bearer genus Coleophora into countless "small genera". That really does not benefit systematics and taxonomy!

If ponteni lies very close to Colias species, I see no advantage in introducing the genus Protocolias, which applies to (its type species) imperialis, and would then also have to apply to ponteni. Only if there had been hundreds or thousands of species would there perhaps be reason to introduce new genera. To introduce a subgenus, on the other hand, can in certain cases be justified, but it should then primarily serve a sensible purpose. I am an advocate for "species groups", which is an informal subdivision of species, where similarities should be the guiding principle. There is then great freedom to place species within the respective group, and later change the placement if new data come to light. No genera need be described that require more formal treatment.