What are todays biggest challenges for a better knowledge of Killifish (oviparous Cyprinodontiformes)?


Dr Jean H. Huber*


* Museum national d'Histoire naturelle, 43 rue Cuvier, 75231 Paris Cedex 05, France.



Techniques and approaches in Systematics have strongly evolved in recent years for fishes and notably Killifishes (Pisces: oviparous Cyprinodontiformes) follow these trends. Thanks to computer software, researchers have become more analytical, bringing more sharply defined or extracted data from their observations. This concerns not only molecular techniques (DNA, RNA, from nucleus or mitochondria) obviously, but also morphology, osteology and lately behaviour. As a result of the availability of more numerous and sharpened data, systematics tend to split more and more, with the limit of differentiation being at the level of human eye, at least within the Linnaeus binominal nomenclature. To progress towards this asymptotic limit, 33 "biggest challenges" in terms of systematics are listed that remain unresolved for older names and an urgent call is given to win another challenge concerning "missing comparative diagnosis". Finally another type of challenge is proposed for action to expert and/or curious aquarists, according to 4 perspectives of cooperative contributions: documentary, collecting trips, breeding and behaviour.



I. Introduction


Techniques and approaches in Systematics have strongly evolved in recent years in biology. Fishes and notably Killifishes follow these trends.

Historically the first change towards more clear-cut criteria began with cladistics, where any character is given a primitive status or a derived status, while only derived characters are considered to group related taxa with synapomorphies (common derived characters).

Thanks to computer software, such as PAUP, TNT, Hennig86, PHYLIP or NONA, researchers have become more analytical, bringing more sharply defined or resolved data from their observations. And the constraint of only considering derived characters has been levied while phylogenetic trees are computed no matter how is assigned each character (even in recent years software were able to compute with more than 2 states of characters, not only '0' for primitive, or '1' for derived, i.e. up to 5 states, plus an uncertain state labelled as '?'). Besides the computation became more and more easy and quick with processors speeding over 1 GHz and the matrix of characters could encompass several hundreds of them.

While the first computations were mainly addressed with limited osteological data -the backbone of cladistics- the technique today boosts not only molecular techniques (DNA, RNA, from nucleus or mitochondria) obviously (with at least 400 genetic characters per matrix), but also independently morphology, osteology and lately behaviour.

Today a matrix of a hundred morphological characters for members of a genus is not unachievable. We were even able to go beyond for a larger scope on upper level groups and superspecies [Huber, J.H. 1998d. A Comparison of Old World and New World Tropical Cyprinodonts. A parallel Outlook of similar and distinctive Characteristics regarding Distribution, Evolution, Ecology, Behavior, Morphomeristics. Soc. fr. Ichtyologie Ed., Paris (Oct. 10): 109 pp., 17 figs.].

However soon problems became apparent and the biological complexity of Killifish ironically jumped again on those who felt that computers and molecular technology were there for an ultimate answer. First, character states cannot be considered in absolute terms, but relatively to each analysed clade. Second, the main problem with the external morphology approach (and also with the osteological and molecular approaches) concerns frequent homoplasies (convergence and reversion), i.e. "faked" data. Third, the major objective of phylogenetic reconstruction -to recognize synapomorphies at nodes of each group- often meets proper difficulties with Killifish, because of their morphological uniformity and because of the limited options combined ad libitum by these fishes (and molecular data did not solve the issue).

Anyway, as a result of the availability of more numerous and sharpened data, systematics tend to split more and more, with the limit being at the level of human eye, at least within the Linnaean universe (binominal names and types in Museum Institutions).

And conversely the need for new data (= new characters, or redefined characters, or new states) has become overwhelming to feed matrixes and computer sessions, and produce new phylogenetic trees and iteratively new knowledge and new unresolved issues.

This is the whole story of necessary data (brand new, updated, or missing data) and obviously of Killi-Data online with the important additional remark that nobody is today able to produce alone enough new data (new characters, new states) to feed the matrixes and the computer sessions. Hence cooperation is imperious ! This means cooperation between everybody, both serious and curious, between professional researchers (who have the educational format) and amateur researchers (who have the dedication) and aquarists (who have time to observe their fish alive, passion and funds to go collecting). Sharing is becoming the master word and aquarists within Killifish Association fit perfectly into that frame, obviously.

To progress towards this asymptotic limit of knowledge fulfilment, 33 "biggest challenges" in terms of systematics are listed that remain unresolved for older names and an urgent call -the diagnostic challenges- is given concerning "missing comparative diagnosis". Finally other challenges -the aquaristic challenges- are proposed for action to expert and/or curious aquarists, according to 4 perspectives of cooperative contributions : documentary, collecting trips, breeding experiments and behaviour.

Future success, if we are able to develop more and more the potentials of the cooperative platform that was created by Killi-Data online, is at our thresholds.




II. Systematic challenges


What are, according to the author, the 33 biggest systematic challenges regarding knowledge of Killifishes ?

1. Rivulus micropus

2. "Cyprinodon" martae

3. Fundulopanchax spoorenbergi

4. Pachypanchax nuchimaculatus

5. Rivulus xanthonotus and a final status for Aphyosemion trilineatum

6. schreitmuelleri : Megalebias vs. Austrolebias

7. Cynolebias porosus

8. Laciris pelagica and Aphanius apodus

9. Pantanodon madagascariensis and Millerichthys robustus

10. Rivulus obscurus andornatus

11. Aphyosemion bualanum

12. Aphyosemion escherichi vs. A. microphtalmum issue (linked to Plataplochilus ngaensis)

13. Fundulopanchax deltaensis, gularis, fallax, kribianus, schwoiseri

14. Fundulopanchax walkeri and/or spurrelli

15. Fundulopanchax powelli

16. viviparous Epiplatys bifasciatus/spilargyreius

17. Epiplatys lokoensis

18. Nothobranchius mkuziensis, orthonotus and rubroreticulatus

19. Poropanchax normani and the Angolan lampeyes

20. Rivulus holmiae andlanceolatus

21. Aphyosemion elegans and decorsei

22. Aphyosemion splendidum, batesii, kunzi

23. Hylopanchax silvestris and stictopleuron

24. Epiplatys nigricans and chevalieri

25. Aphyosemion ferranti, lujae, Epiplatys multifasciatus

26. Pterolebias bokermanni, luelingi and the rediscovery of longipinnis

27. melantereon : Scriptaphyosemion vs. Epiplatys

28. Lacustricola atripinna and bukobanus

29. Fundulus kansae and zebrinus

30. Aphyosemion exiguum and Epiplatys nyongensis

31. Some disturbing Aplocheilus issues for blockii, panchax, siamensis, andamanicus

32. The unsatisfactory situation of Orestias, intralacustrine speciation or not

33. The numerous names with missing types or undisclosed type material


These 33 systematic challenges are all thorny questions that remain in front of us, and now that the cooperative community of Killi-Data is a fact, not an idealistic dream, it is achievable… provided of course that political conditions in the concerned country are stable and health issues are secured, in case of needed new fish collections.


* 1- Rivulus micropus : since Huber [1992. Review of Rivulus. Ecobiogeography - Relationships. Cybium Suppl., Société Française d'Ichtyologie Publ.: 586 pp., 40 pls., 85 figs., 8 tabs, 13 maps], the status of that very old name (Steindachner, 1863), even older than urophthalmus described in 1866, is a major problem ; its type locality is so imprecise ("Rio Negro", a large river, more than 2000 kilometres long) and the morphology of the single type is so average (actually close to urophthalmus/rubrolineatus or to limoncochae/iridescens) that it seems a desperate case ; only, the study of old manuscripts at the Vienna Museum or the analysis of the specimens that were collected on the same day (by Natterer, in 1830 !) might provide with a clue regarding the exact type locality … but then how to decide for a senior synonymy with urophthalmus/rubrolineatus (precisely, compressus has been described from the surroundings of Manaus) or with the fish with half the number of lines near Caudal peduncle (referable to aff. limoncochae) available in the Rio Negro, toounless a consensus is raised among Cyprinodontiformes researchers to request ICZN to suppress the taxon micropus

* 2- "Cyprinodon" martae : this question is currently under study by the author, after a first publication in 2000 [On nomina oblita among Cyprinodont Species. J. Amer. Killifish Assoc., 33 (2): 43-51] ; the single type clearly does not correspond to a Cyprinodon fish, but instead, it may be an annual form belonging to a new distinctive genus, according to the photograph. Pending the actual report on the study of that specimen by the author, the challenge lies in rediscovering the living fish near Santa Marta {11.250N;74.200W}, near the mouth of the Río Magdalena, in northeastern Colombia and at the same time, collecting fish corresponding to 2 names that may have been considered synonyms too quickly : Austrofundulus myersi and Rachovia splendens (both described by Dahl from Sincelejo {9.420N;75.720W}, northern Colombia) ; and, parallely, in assigning a precise type locality for the 2 older valid names, Austrofundulus transilis from Guarico State (without details), Orinoco basin, Venezuela and Gnatholebias zonatus from Guarito county (without details), Orinoco basin, Venezuela …

* 3- Fundulopanchax spoorenbergi : this case is apparently very simple, the name was described after an import from unknown origin (probably near the boundary between southeastern Nigeria and northwestern Cameroon, between Calabar and Mamfé) and the challenge is restricted to re-discover the fish ; however the original import brought 2 colour phases, apparently sympatric fishes, the other one being more closely related to a gardneri type [see Wildekamp, R.H. 1996. A World of Killies. Atlas of the Oviparous Cyprinodontiform Fishes of the World. Vol. 3. Amer. Killifish Assoc. Publ.: 330pp, figs.] ; if the sympatry is confirmed, then are there 2 distinct species or is it only polymorphism, and how should be re-evaluated the case of Fp. obuduensis [Wright, F. & J. Jeremy. 1974. Aphyosemion gardneri obuduense. A Description of a new Aphyosemion gardneri Subspecies from Nigeria. British Killifish Ass. Publ., Separatum, 103: 4 pp., fig.]? Another reason for in-depth collections in southeastern Nigeria and around…

* 4- Pachypanchax nuchimaculatus : a difficult case, because the single type, rather distinctive [Huber, J.H. 1998a. Miscellaneous Notes on some Systematic Difficulties Regarding old World Cyprinodonts. J. Amer. Killifish Assoc., 31 (1): 3-17, 28-32] is from an unknown locality in Madagascar that is not a small island ; however the diagnostic character (at the origin of the name : marking on neck) and the recent knowledge that Pachypanchax is not present in the southern part of the island open the possibility of solutions, with necessary in-depth collections…

* 5- Rivulus xanthonotus and a final status for Aphyosemion trilineatum : these 2 cases are rather similar, although they concern very distant fishes ; both fishes have been described from aquarium imports without certain origins, but while the former has since long been considered as valid (with a lectotype in Berlin Museum), the latter is not seen the same by all researchers ; Rivulus xanthonotus has its type locality as "Amazonas", however Hoedeman has restricted it to Obidos (but without evidence), near Santarem, lower Amazon, Brazil, i.e., not far from the type locality of Pterolebias longipinnis, another difficult case ; Aphyosemion trilineatum has its type locality in "Cameroon" (without details) ; Lazara, K.J. [1984. Killifish Master Index. 3rd Edition. Amer. Killifish Assoc. Publ.: 295 pp.] considers it as a possible junior synonym of Aphyosemion cameronense, while Wildekamp [1993a. A World of Killies. Atlas of the Oviparous Cyprinodontiform Fishes of the World. Vol. 1. Amer. Killifish Assoc. Publ.: 311pp, figs.] considers it as a nomen dubium (a doubtful name) ; to tackle again the issue means analysing closely the German aquarium magazines and German Aquarium Associations leaflets between 1920 and 1935 and pinpoint any new data : a frustrating, but necessary assignment …

* 6- schreitmuelleri : Megalebias vs. Austrolebias : this case is not simple at all, because again, authors are not in agreement ; the type locality, Rio de Janeiro, Brazil, may be erroneous because it is again a trade import (but lately Leptolebias marmoratus was rediscovered live from the same region after 50 years !) ; this single aquarium import of three males, ever, is variably analysed: either as a synonym of Megalebias wolterstorffi (then from Uruguay, not Brazil, as per Lazara, K.J. [1982b. The Taxonomy and Nomenclature of South and Central American Killifishes. J. Amer. Killifish Assoc., 15 (5): 166-184] or as an Austrolebias valid species that needs recollecting [as per Wildekamp, R.H. 1995b. A World of Killies. Atlas of the Oviparous Cyprinodontiform Fishes of the World. Vol. 2. Amer. Killifish Assoc. Publ.: 384pp, figs.], or finally as a nomen dubium (a doubtful name) also in Austrolebias [as per Seegers, L. 2001a. Killifishes of the World. New World Killis. III. Aqualog, Mörfenfelder-Walldorf, Band 12: 210 pp., figs.] ; all 3 authors agree on one issue : the fish does not originate from Rio in Brazil, but from Uruguay or the neighbouring region of southeastern Brazil, i.e. very far from Rio ; since the 3 types appear to be lost in Berlin Museum, the situation is not brilliant …

* 7- Cynolebias porosus  : this case is important because, like for Rivulus micropus, it is a very old taxon described by Steindachner, in 1876 ; shipped from Pernambuco (today Recife; in fact from inland of Pernambuco state, but probably not far, due to travel contingencies), in northeastern Brazil, Costa [2001, The neotropical annual fish genus Cynolebias (Cyprinodontiformes: Rivulidae): phylogenetic Relationship, taxonomic Revision and Biogeography. Ichthyol. ExpIor. Freshwaters, 12 (4): 333-383, 32 figs., 9 tabs.], hypothetically, suggests that it could have been collected during (or brought to) the Thayer (North American) expedition (Louis Agassiz), on July 31. or August 1. 1865 ; these clues should allow to rediscover the fish if it is not extinct because the region is extremely dried out today and urbanization has dramatically increased within a century ; but let's not be pessimistic : Cynolebias microphthalmus, that Costa sees as distinct, is still living not far …

* 8- Laciris pelagica and Aphanius apodus : both cases are similar because these 2 fishes are extremely endangered, if not extinct ; both species are unique : for pelagica, it is endemic to lake Edward in the Rift Valley of central Africa (deep waters of Lake Edouard, Zaïre and Uganda {0.350S;29.583E}) ; for apodus, it is the single, very old Aphanius species without ventral fins (in all known localities, it is extinct : the only solution is to discover a brand new locality in Algeria) ; the problem is Man who introduced alien species that phase them out ; shame on us …

* 9- Pantanodon madagascariensis and Millerichthys robustus : both cases are similar because these 2 fishes are extremely important in the understanding of Killifish phylogeny (both are probably very primitive because they belong to very primitive areas) ; the problem is not their type localities that are not too difficult to reach (madagascariensis : Mahambo, eastern coast, Madagascar {17.260S;49.230E} ; robustus : 5 km S. Papaloapan (today El Hule), Mexico {18.170N;96.100W}) ; the problem is again Man who made these species extremely endangered, if not extinct, because he drastically changed their very reclusive biotopes to develop better his own conditions of life ; who will face the challenge (with collecting permits) ? Shame on us (again)…

* 10- Rivulus obscurus and ornatus : this is a complicated case and a confusion due to the simultaneous publications of 2 authors (end of 1992 and beginning of 1993, without knowing the project of each other): first, the author's Rivulus book [Huber, J.H. 1992. Review of Rivulus. Ecobiogeography - Relationships. Cybium Suppl., Société Française d'Ichtyologie Publ.: 586 pp., 40 pls., 85 figs., 8 tabs, 13 maps.] where lectotypes of ornatus and obscurus are designated ; and second, Costa's paper in DKG Journal [Costa, W.J.E.M. 1993. Zur Identität und Verbreitung von Rivulus ornatus und Rivulus punctatus. D.K.G. (Deutsche Killifisch Gem.) J., 25 (3): 44-46, 2 figs.] where he mentions that the true ornatus is in fact obscurus according to material from near Manaus (3.130S, 60.020W) and the aquarium ornatus is then an unknown species, without knowing the lectotype designation for ornatus from Silves, i.e. not in the area of Manaus ; this is unfortunate and lies only in the lack of communication between Costa and Huber ; but the problem actually comes from the fact that in his descriptions, Garman (1895) used several lots for ornatus with 3 different and distant localities and one lot for obscurus with a type locality near Manaus, identical with one of the 3 for ornatus : in details for ornatus, the 3 localities are Silves, Lake Saraca, N. Brazil {2.880S, 58.350W} ; Parana do Janauari, Brazil {3.200S, 60.080W} (near Manaus); Lago Cudajas, now L. Badajos, N. Brazil {3.250S, 62.780W} and for obscurus, Lago Januaria, vicinity of Manaus, N. Brazil {3.200S, 60.080W} ; by selecting Silves as the official type locality for ornatus, without knowing Costa's research, Huber made Costa's publication inappropriate because Costa allocated ornatus to a fish we do not know if it is identical with the fish from Silves ; let's wait until live material is collected from the 2 fixed type localities, Silves for ornatus, and Januaria, near Manaus for obscurus and the issue can be fixed. If the 2 fish are identical, then (ICZN law of first reviser) ornatus has priority over obscurus ; if they are distinctive, then the 2 names remain valid, which, in turn, will enable to address another issue, the identification of the aquarium strains of ornatus : one as an aquarium import from Obidos (?), lower Amazon (NSC-2), Brazil {1.920S, 55.520W} and the other from Padre Isla, Iquitos, Peru {3.620S, 73.700W} …

* 11- Aphyosemion bualanum : this question was raised after Seegers's publication [1988d. Bemerkungen über die Sammlung der Cyprinodontiformes (Pisces: Teleosta) des Zoologischen Museums Berlin. 1. Die Gattungen Aphyosemion Myers, 1924 und Fundulosoma Ahl, 1924. Teil 2. Mitt. Zool. Mus. Berlin, 64 (1): 3-70, figs.] and Huber's reaction [1998: Miscellaneous Notes on some Systematic Difficulties Regarding old World Cyprinodonts. J. Amer. Killifish Assoc., 31 (1): 3-17, 28-32] ; after the study of the same single type of bualanum, Seegers believes that it belongs to an unknown Mesoaphyosemion sp. (e.g. like cameronense or wildekampi) and Huber is cautious and does not find morphological differences with what was understood as bualanum since Scheel (a fasciated fish) ; no solution from the renewed study of the single type : it has been smashed for unknown reasons ; to solve that issue, only successful collections around Bouala {6.417N;15.583E}, in Central African Republic, are required, but this is a challenge : a remote place in highlands (1200 m altitude) with no primary forest …

* 12- Aphyosemion escherichi vs. A. microphtalmum issue (linked to Plataplochilus ngaensis) : this question was also raised after Seegers's publication [1988d. Bemerkungen über die Sammlung der Cyprinodontiformes (Pisces: Teleosta) des Zoologischen Museums Berlin. 1. Die Gattungen Aphyosemion Myers, 1924 und Fundulosoma Ahl, 1924. Teil 2. Mitt. Zool. Mus. Berlin, 64 (1): 3-70, figs.] and Huber's reaction [1998: Miscellaneous Notes on some Systematic Difficulties Regarding old World Cyprinodonts. J. Amer. Killifish Assoc., 31 (1): 3-17, 28-32] ; after the study of the same type series of escherichi, Seegers believes that it is a senior synonym of microphtalmum and Huber is cautious and does not find morphological differences with striatum (a lineated fish that is often sympatric with microphtalmum in northern Gabon) ; to solve that issue, only collections near Attogondema, Nga river, German Cameroon (today Attokondama, tributary to Noja river, northern Gabon, near the border with Mbini, according to Seegers) are required, but this is a challenge : a remote place in the foothills {0.850N;10.250E} of Monts de Cristal highlands with the possible sympatry of the 2 species ; if sympatry is also recorded there, then only a molecular study of the types may solve the problem ; in any case the collection is important for 2 reasons : first, escherichi may be a different fish, related to A. aff. cameronense from the foothills of Monts de Cristal (that Huber has refrained to describe as a new species, in 1977, precisely to avoid any risk of future synonymization) ; second, because it is also the type locality of Plataplochilus ngaensis which is unknown alive (after its rediscovery, a full since long awaited revision of the genus Plataplochilus could be undertaken) …

* 13- Fundulopanchax deltaensis, gularis, fallax, kribianus, schwoiseri : this question was also raised after Seegers's publication (1988d. Bemerkungen über die Sammlung der Cyprinodontiformes (Pisces: Teleosta) des Zoologischen Museums Berlin. 1. Die Gattungen Aphyosemion Myers, 1924 und Fundulosoma Ahl, 1924. Teil 2. Mitt. Zool. Mus. Berlin, 64 (1): 3-70, figs.] and Huber's reaction is to be published soon after a re-evaluation of the case (Seegers proposed the synonymization of kribianus and schwoiseri with the older name fallax, on the basis of material in BMNH and on aquarium imports of the same years in Germany) ; Killi-Data online opts for a conservative situation, considering deltaensis and gularis as valid, kribianus and schwoiseri, as valid too, and fallax as a nomen dubium), pending Huber's publication ; no doubt that this will not be the end of the story ; apart from the renewed study of preserved material in Museums, the target is indeed an in-depth sampling of annual forms of that genus in southern Nigeria ; there is little doubt that the discontinuity in populations of sjoestedti in western Nigeria and in northwestern Cameroon does not reflect reality : between west of the mouth of Niger river and the Cameroon boundary there must be populations of large annual Fundulopanchax species with a median band on sides of both sexes, if not deltaensis

* 14- Fundulopanchax walkeri and/or spurrelli : this case is a very long story and after more than 20 years without new evidence it is probably time to reassess it calmly, considering a third taxon in addition, Fundulopanchax litoriseboris ; all 3 have been considered either as valid species, or valid subspecies of walkeri, or the last 2 taxa as conspecific with walkeri, or a mix of these options, sometimes with a consensus approach, sometimes with a controversial approach ; details in Radda, A.C. [1976c. Description of Aphyosemion litoriseboris spec. nov. J. Amer. Killifish Assoc., 9 (11): 301-305, 2 figs.], then Berkenkamp, H.O. [1976c. Über Fundulus walkeri Boulenger, 1911, Fundulus spurelli Boulenger, 1913 und Aphyosemion litoriseboris Radda, 1976. D.K.G. (Deutsche Killifisch Gem.) J., 8 (12): 151-160, 5 fig.], Berkenkamp, H.O. [1977b. Zum Abschluss der Aphyosemion litoriseboris Diskussion. D.K.G. (Deutsche Killifisch Gem.) J., 9 (8): 121-128, map.], Berkenkamp, H.O. & V. Etzel [1980c. Aquarienfische aus der Elfenbeinküste. 7. Die Formen von Aphyosemion walkeri aus der Elfenbeinküste und Ghana. D.K.G. (Deutsche Killifisch Gem.) J., 12 (3): 33-40, 12 figs.; 12 (4): 51-58, 5 figs., map.], Berkenkamp, H.O. & V. Etzel. 1981a. [Aquarienfische aus der Elfenbeinküste. 8. Kreuzungen mit Formen von Aphyosemion walkeri (Blgr., 1911). D.K.G. (Deutsche Killifisch Gem.) J., 13 (5): 71-81, 16 figs.]; this case is full of scientific weaknesses : the first 2 taxa are unknown live from their type localities in Ghana since their description and the type locality of the third is an aquarium import ; the first duty would then be to fill these "holes" with hard data and live fish ; spurrelli : vicinity of Bibianaha, near Dunkwa, between watersheds of Tano and Ankobra rivers, Ghana {5.167N;2.700W} ; walkeri : Bokitsa Mine (a misspelling for Bokitsi Mine within Wasa district, 1.2 km south Ayenfuri or Ayenfor and 18 km from Dunkwa, i.e. about 55 km south-southwest of Kumasi, subseq.), Ghana {5.983N;1.900W} ; and litoriseboris : aquarium strain collected by Arnoult in 1963 (bred by E. Pürzl), from Ivory Coast (should not be too difficult to precise from Arnoult's notes or old Aquarium magazines) ; a patient, objective work …

* 15- Fundulopanchax powelli : this case is apparently very simple, the name was described from a precise locality (N.W. Bakakodia, Nana creek, ca. 20 km northeast of Escravos river mouth and 55 km west northwest of Warri, northwest of Bakakodia or Okokodiagbne village, Delta state, Niger delta, S. Nigeria {5.672N;5.325E}) ; however, only preserved juveniles are known and first attempts to re-collect the fish, there, did not succeed ; knowing the dynamisms of fish exporters in Lagos, active correspondence and money, at least, should be a spur to deliver a solution for this very special fish, probably primitive and a key to the understanding of the huge speciation in the Niger delta…

* 16- viviparous Epiplatys bifasciatus/spilargyreius : since Guma'a [1982. On the Biology of Female Epiplatys bifasciatus (Cyprinodontidae) from Southern Sudan. Hydrobiologia, 89: 285-300, 5 figs., 8 pls., 3 tabs.] and since the description of Epiplatys marnoi [Steindachner, F. 1881a. Ichthyologische Beiträge (X). Sitz. Akad. Wiss. Wien mathem.-naturwiss. Klasse Abt.1, 83: 179-219, 8 pls. (17 figs.)], there are strong suspicions that a viviparous Epiplatys species is available in eastern Africa ; it is unknown if it will be identifiable to bifasciatus, or to marnoi (a present synonym of spilargyreius), but it is a fascinating venture…

* 17- Epiplatys lokoensis : this case is also apparently simple like powelli ; recollections of the fish in the very limited area of Sierra Leone (Bonkorkon, west of Port Loko, Port Loko district) should allow to re-study the fish and disclose if the describers were right to erect a new name or if the synonymization by Romand [1980b. Comments on Epiplatys lokoensis Berkenkamp & Etzel (1978) from Sierra Leone (Pisces, Cyprinodontidae). Rev. Zool. Bot. Afr., 94 (3): 591-599, 9 figs., tab.] with barmoiensis was correct ; but the whole question of the bifasciatus superspecies should be re-evaluated because the separation by Romand into 2 valid subspecies (with bif. bifasciatus and bif. taeniatus), based on habitat (respectively savannah vs. forest) is discussed, because the type locality of taeniatus is not in forest (Jebba, Niger river, N. Nigeria {9.117N;4.817E} Madsen, 1927) and should be restudied with a comparative diagnosis ; and first of all, a precise type locality and the first (ever) collection of live topotypes of Ep. bifasciatus (type localities : Bahr-el-Seraf (Zeraf) and Bahr-el-Gebel (Jebel), White Nile system, Sudan) are prerequisite …

* 18- Nothobranchius mkuziensis, orthonotus and rubroreticulatus : since a World of Killies [Vol. IV, AKA Publication, 2004, also as a pers. comm. to Killi-Data online], Wildekamp has proposed that the status needs a re-evaluation ; it may be a distinct species related to rachovii, anyhow clearly not a junior synonym of orthonotus as previously thought ; a re-discovery is thus highly interesting at type locality : Mkuzi (today Mkuze) river, Natal (province), South Africa {27.417S;32.417E}; besides, the first live collection of Nothobranchius rubroreticulatus (a species described already years ago) would be also desirable to separate it also from isomorphic populations that Bellemans discovered in Sudan ; the type locality : near the capital city of Chad (Koundoul, S. N'Djamena, S.W. Tchad {11.960N;15.150E}) ; these "easy" challenges must not hide the critical issue remaining in Nothobranchius : are allopatric colour variations possibly distinct species, and notably is mkuziensis distinct from rachovii or, around Beira in Mozambique, kuhntae distinct from orthonotus ? Not the end of the story…

* 19- Poropanchax normani and the Angolan lampeyes : this case is an old one, since Poropanchax normani, and Lacustricola mediolateralis, Lacus. nigrolateralis, Lacus. (?) macrurus have never been collected live from their type localities and their descriptions are more than 50 years old ; this is notably a pity for normani (is it a true Poropanchax ?) and macrurus (is it related to spilauchen or not at all ?) that are old taxa (respectively 1928 and 1904) and important phylogenetic units ; type localities are : normani, near Katagum, Kiyawe river, N. Nigeria {12.283N;10.350E} ; macrurus, Marimba, Lake Sarmento (near mouth of Cuanza river, subseq.), N.W. Angola {8.367S;17.033E} ; mediolateralis, near Cuango-Muquehe river source, N.E.Angola {10.850S;19.300E} ; nigrolateralis, Tchimenji river, N.E. Angola {7.967S;21.117E} ; targets for lampeye lovers …

* 20- Rivulus holmiae and lanceolatus : both cases correspond to fish described by Eigenmann in 1909-1912 that apparently cannot be re-collected despite major efforts in recent years ; the description of Riv. holmiae is very good for the beginning of the twentieth century, plus numerous type specimens of both sexes are available in good conditions ; at type locality (near Holmia, at entrance of Chenapowu river, upper Potaro river basin, Guyana (ex British Guiana), 500 m altitude {4.970N;59.580W} ), only Riv. waimacui could be caught ; by morphology, holmiae is close to igneus and to immaculatus ; it differs from igneus by the frontal scalation, the male colour pattern and minor morphological characters ; it does not differ from immaculatus except by having a supracaudal ocellus in female (but this is a variable character, like the dark border at male Caudal) ; in addition, holmiae and immaculatus are living in the plateau ; of course both may well not be synonyms, but our thinking is biased by the historical collections in the lowlands of Guyana and Suriname that were identified as holmiae in the sixties and that may be closer to hartii (or to igneus) ; on the other hand, the description of lanceolatus is less obvious and a single type, in poor condition, is available : the unique feature is the "lanceolate" shape of the caudal fin (that cannot be confirmed from the type) ; at type locality (Rockstone, (British) Guiana {5.980N;58.560W} ), only stagnatus-like fish have been collected, yet ; various hypothesis have circulated, including a possible identity as a juvenile Moema sp. that is also reported in the area (apparently ruled out), a possible synonymy with stagnatus (unlikely), a possible synonymy with agilae (then a female specimen) ; no precise argumentation can be forwarded to demonstrate that Eigenmann made a mistake about the type localities or in the descriptions : he stayed in the region several months and as a researcher his reputation is excellent ; then to collect intensively again in the type area of holmiae (or in the region in-between with immaculatus) and to compare the material with the lowlands in Guyana, for the first case, and in the type area of lanceolatus, for the second case, are the only way for better times …

* 21- Aphyosemion elegans and decorsei : this case is an old one and concerns the Congo cuvette like the 4 next cases ; elegans and decorsei are the oldest members of the lyre-tail Aphyosemion group, mainly dwelling that huge basin ; the former has been described in 1899 and the latter in 1904 ; not surprisingly both are unknown live from their type localities (Bikoro, Lac Tumba {0.750S;18.117E} and Mbandaka {0.067N;18.267}, N.W. Zaïre for elegans and Bessou, Oubanghi basin, S.E. Centrafrique {5.090N;19.540E} for decorsei) ; fish have been assigned to these 2 names : a fasciated fish common not far the type locality for elegans and a poorly (or strongly, depending on authors !) punctuated fish for decorsei ; these identifications may be right or completely wrong ; for example, first, in his description of elegans, Boulenger does not mention the bars on male sides, second in sampled regions of Central African republic for decorsei, there are at least 2 very distinct phenotypes (with many spots and with few spots) and third, 2 sympatric species of that group are not rare in general in the cuvette ; then, who is who, a key issue to correctly identify these 2 species and all other (about 10 !) members of that superspecies subsequently described from the cuvette (including topotypes of schoutedeni at Madié: 2.417N; 27.302E)…

* 22- Aphyosemion splendidum, batesii, kunzi : this case is confusing since taxonomic and systematic issues are mixed ; first, the cuvette dweller, splendidum, has never been studied alive (material has been collected live from the type locality, near Ouesso {1.617N;16.067E} in northern Congo and several other spots in the cuvette) but it was not brought back alive in labs : the phenotype (poorly dotted on sides and fins) is distinct from batesii from central Cameroon (the latter being deeper bodied, with many large spots and no chevrons on sides and large and equal yellow margins on fins of male) and from kunzi (with anterior chevrons and a lower band on sides and unequal yellow margins on fins of male) ; is it polymorphism (conspecific variability) as Brosset [1982. Le Peuplement de Cyprinodontes du Bassin de l'Ivindo, Gabon. Rev. Ecol. (Terre Vie), 36: 233-292] has shown from in-depth samples in Gabon or is the picture much more complex (with 2 or 3 species being valid) as aquarists tend to believe ? Only many DNA samples may allow a better understanding of these relict fishes that are semi-annuals, but not close to Fundulopanchax (to what then ?)…

* 23- Hylopanchax silvestris and stictopleuron : this case is a difficult one, like many systematic issues concerning lampeyes ; since Huber's collections [1982a. Rapport de Synthèse sur l'Expédition au Congo (1978). Cyprinodontidés récoltés et Micropanchax silvestris, Synonyme de stictopleuron. Rev. fr. Aquariol. Herpétol., 9: 1-12, 9 figs., tab., 5 maps], not far from the type locality of stictopleuron (Oka, 18 miles (29 km) north of Eovo, Congo basin {3.600S;15.267E}), both names have been synonymized ; however, Lazara [2001. The Killifishes, an annotated Checklist, Synonymy and Bibliography of recent Oviparous Cyprinodontiformes Fishes. Killifish Master Index. 4th Edition. Amer. Killifish Assoc. Publ.: 624 pp., 3 appendixes], after the study of several Museum samples from Zaïre (silvestris type locality : Yangambi, Lusambila river {0.783N;24.400E}), believes that the morphological variation is so big that the 2 names must be valid ; then 1, 2 or more valid species for this phenotype dwelling the entire Congo cuvette ? Again, only many DNA samples may allow a better understanding of these relict fishes that pushed the Hypsopanchax phenotypes (with several valid names !) to the periphery of the cuvette …

* 24- Epiplatys nigricans and chevalieri : this case is similar to Hylopanchax ; since Huber's collections all along the western belt of the Congo cuvette [1979b. Cyprinodontidés de la Cuvette Congolaise, Adamas formosus n. gen., n. sp. et nouvelle Description de Aphyosemion splendidum. Rev. fr. Aquariol. Herpétol., 6: 5-10, 6 figs., 2 tabs.], both names have been synonymized and this was not a big surprise : several authors had expressed doubts and difficulties in separating them, and concluded either they could be synonyms or subspecies ; expert aquarists, though, did not follow this stance ; up to now no new published evidence has been put forward to change the field results ; Wildekamp [1996. A World of Killies. Atlas of the Oviparous Cyprinodontiform Fishes of the World. Vol. 3. Amer. Killifish Assoc. Publ.: 330pp, figs.] has proposed that western populations (i.e. west of Congo river) be named chevalieri while eastern ones be named nigricans, on the basis of colour pattern differentiation of some populations ; but that was tentative to maintain a conservative position and he did not used the same model for the multifasciatus group (boulengeri / multifasciatus) with strong genetic gaps, though ; indeed, there are differences between populations of chevalieri/nigricans from the Congo cuvette, but it is simply unknown if these differences are stable, if they correspond to strong genetic divergence, if they are linked to geographical criteria (there is absolutely no barrier, but the concept of refugium may apply, like for lake Tumba) ; all is known are a few populations from the Kisangani area (nigricans type locality : Dungu, at the Kibali-Dungu river junction, northeastern Zaïre {3.633N;28.567E}), from the Brazza area ({4.267S;15.250E}, type locality of chevalieri), from the Likouala area and from isolated spots like the one from lake Fwa, thousands of kilometres apart ; and neither molecular studies, nor detailed morphological measurements have been performed on these populations ; it is totally insufficient…

* 25- Aphyosemion ferranti, A. lujae, Epiplatys multifasciatus  : this is an old case since the 3 names were described from the same locality (near Kondué, Kasai, Congo (today Zaïre) {4.983S;23.300E}; to collect there would allow a proper identification of the 3 phenotypes, an assignment to a phylogenetic group for ferranti and a new detailed study of the multifasciatus superspecies ; and by the way, another series of live discovery is to be found not "far" for the first time : Hypsopanchax jubbi (Near Zambezi source, Mwinilunga, Zambia {11.117S;24.117E}) and H. jobaerti (Lula, Mosanji river, S.W. Zaïre {7.217S;23.117E}), plus Nothobranchius species, N. (Zono.) brieni (Bukama, Shaba Province, S.E. Zaïre {9.208S;25.850E}), N. (Zono.) malaissei (1.5 km E. Kabiashia, S. Lake Moero, S.E. Zaïre {10.267S;28.133E}), N. (Zono.) polli (Near Mwadingusha, Shaba, S.E. Zaïre {10.750S;27.250E}) ; really an un-deserved situation …

* 26- Pterolebias bokermanni, luelingi and the rediscovery of longipinnis : indeed a strange case that shows how things may change unexpectedly ; since Thomerson [1984. Rivulichthys luelingi, a junior Synonym of Pterolebias longipinnis (Pisces: Rivulidae). Copeia, (2): 528-529, fig.] the status of luelingi is fixed as a synonym ; and since Costa [1988a. A new Species of the neotropical annual Fish Genus Pterolebias (Cyprinodontiformes, Rivulidae), from Brazil. J. Zool. Soc. London, 215: 657-662, fig.] bokermanni is also fixed as a synonym of longipinnis ; however Staeck [1994b. Die Killifische Amazoniens. Aquar. Terr. Zeit. (D.A.T.Z.), 47 (11): 692-696, figs.], Huber (1995), Lazara (2001), Costa (2003), following collections in Brazil, Paraguay, Bolivia and Argentina and the observed variations in morphology and colour pattern, began to suspect that there may be more than one species there ; bokermanni, described in 1955, from Rio Guajara-Mirim, Guaporé, W. Brazil {10.910S;65.150W} and luelingi described in 1969, from Rio Chapare, PK4 Todos Santos, Bolivia {16.830S;65.170W} might not be synonyms each other (bokermanni is anyway the older name) and above all, may well be distinct from the unknown live and very distant longipinnis ; first step to prove it : the rediscovery of longipinnis at Santarem, Amazon river, Pará (state), Brazil {2.420S;54.730W}, with as a bonus the simultaneous re-discovery of Rivulus xanthonotus at Obidos, near Santarem ; a fascinating DNA sampling and morphological diagnosis are on the agenda, from lower Amazon to northern Argentina, over thousands of kilometres …

* 27- melantereon : Scriptaphyosemion vs. Epiplatys : this case is old and un-settled ; melantereon has been described by Fowler [1950. Some Fishes from Liberia, West Africa, with Descriptions of two new Species. Notulae Naturae, 225: 8 pp., 2 pls., 11 figs.] from Robertsport, Liberia, collected by Charles R. Matlock Jr (1946) together with the types of matlocki, an acknowledged synonym of Epiplatys fasciolatus (and also coastal species such as Pseudepiplatys annulatus, Poropanchax normani, Aplocheilichthys spilauchen ; the study of the types (juveniles) of melantereon, reported in Huber [1978. Contribution à la Connaissance des Cyprinodontidés de l'Afrique Occidentale: Caractères taxonomiques et Tentative de Groupement des Espèces du genre Aphyosemion (Cyprinodontidés). Rev. fr. Aquariol. Herpétol., 5: 1-29, 39 figs., 6 maps.], pushed to position the taxon in Scriptaphyosemion and then by zoogeography as a junior synonym of liberiense, and discards its placement in Epiplatys ; however, Wildekamp [1996. A World of Killies. Atlas of the Oviparous Cyprinodontiform Fishes of the World. Vol. 3. Amer. Killifish Assoc. Publ.: 330pp, figs.] still places the name in Epiplatys, as a nomen dubium (a doubtful name) ; then, the renewed study of the material in Philadelphia museum would be the simple answer, easy if only the types were fully adult fish…

* 28- Lacustricola atripinna and bukobanus : this is a forgotten, but important case (bukobanus is the type species of the subgenus Cynopanchax) ; since Wildekamp [1995b. A World of Killies. Atlas of the Oviparous Cyprinodontiform Fishes of the World. Vol. 2. Amer. Killifish Assoc. Publ.: 384pp, figs], Lacustricola atripinna (ex-Aplocheilichthys) described from Busisi, southern banks of lake Victoria, Tanzania {2.733S;32.867E}) is hypothesized to be a senior synonym from bukobanus, from Bukoba, also in Tanzania {1.333S;31.817E} ; a simple, but unavoidable comparative study of the types of the 2 taxa, both in Berlin Museum, is a prerequisite before in-depth live collections…

* 29- Fundulus kansae and zebrinus : this case is recent and emerged from the DNA evidence that the fish populations over the huge range may correspond to 2 distinct species (northern populations assigned to kansae, southern populations to zebrinus) ; both names have no serious type locality, though : kansae , Kansas state (without details), U.S.A. and zebrinus, between Fort Defiance (35.75N;109.11W) and Fort Union (35.92N;105.03W), (eastern) New Mexico state (probably in an upper tributary of Rio Grande del Norte or of Rio Brazos), U.S.A. (by the way, let's not forget the similar case of sciadicus, with Platte river, Nebraska, U.S.A., a very long river indeed, for type locality); hence, there is an urgent need for a proper systematic re-definition of these 2 taxa of the subgenus Plancterus ; it requires "old fashioned" ichthyologists who can carefully study very old publications and museum files in the USA and all are aging now…

* 30- Aphyosemion exiguum and Epiplatys nyongensis : this case would have been anecdotal, only, a few years ago ; both names are stable, systematically speaking : exiguum is a well known valid species, and nyongensis is an acknowledged junior synonym of sangmelinensis ; the type locality (Nyong river, without detail) is identically labelled for A. exiguum and for Ep. nyongensis, but the collectors are not the same and the collections are not dated from the same year ; Wildekamp [1993a. A World of Killies. Atlas of the Oviparous Cyprinodontiform Fishes of the World. Vol. 1. Amer. Killifish Assoc. Publ.: 311pp, figs.] did give geographical coordinates for the type locality of exiguum, but these correspond to the mouth of Nyong river into the Atlantic ocean : a very minor error, but unfortunate for this plateau-dwelling species ; regarding nyongensis, it may end up into a problem too ; Amiet has collected a strange distinctive Epiplatys species [pers. comm. and Vandersmissen, J.P. 2003. Le Groupe Epiplatys sangmelinensis. Assoc. Killiphile Francophone de Belgique, Killi Contact, 30 (5-6): 1-24, figs, map.] that is less deep than sangmelinensis, with some resemblance to the bifasciatus-chevalieri superspecies at Andjeck, not far from the upper Lobo river, the type locality of sangmelinensis : the single type of both nyongensis and sangmelinensis should then be re-studied to clarify this point ; another difficult (not anecdotal any more !) task for "old fashioned" (also aging) ichthyologists, but this time in London and Berlin and a necessary comparison with topotypic live material …

* 31- Some disturbing Aplocheilus issues, blockii, panchax, siamensis, andamanicus : the taxonomic situation is indeed below standard ; most names have been described during the second half of the 19th century probably from colonial harbours (but without certainty) or from aquarium imports without precise origin (and often types are missing in addition); a dozen of names, most probably synonyms, then the stake is not major (rubrostigma, vittatus, buchanani, chrysostigmus, kuhlii, melanotopterus, melastigmus, etc.) ; the issue should at least be tackled for valid or maybe valid names, though : Aplocheilus panchax, the oldest taxon (1822) with a complete re-definition and a precise type locality (Bengal, Ganges river basin, without details, India); Aplocheilus blockii, with a neotype to be designated from Cochin, Kerala, India, Aplocheilus panchax andamanicus, with the first live discovery and a neotype to be designated from Port Blair, Andaman Island, off Burma, India {11.670N;92.730E} ; Aplocheilus panchax siamensis with the disclosure of the precise origin (somewhere in Thailand of this distinctively patterned population) and a renewed live collection …

* 32- The unsatisfactory situation of Orestias, intra-lacustrine speciation or not : this case is indeed an undeserved picture, with the genus Orestias being neglected despite its very high interest (the parallel situation of the Rift Valley Cichlids is strikingly opposite) ; the issue for these Andean fishes that have been uplifted by 4000 meters with the upheaval of the South American Cordillera, some 30 million years ago is : speciation (like in the Rift Valley lakes) or not ? According to Parenti's morphological and osteological study |1984a. A taxonomic Revision of the Andean Killifish Genus Orestias (Cyprinodontiformes, Cyprinodontidae). Bull. Amer. Mus. Nat. Hist., 178 (2): 109-214, figs.], there are over 40 species, most of them being endemic to lake Titicaca ; while according to crossings, morphological virus-induced variations and pattern variability, there are only 4 to 6 valid species in Villwock [1986. Speciation and adaptative Radiation in Andean Orestias Fishes. In: Vuilleumier, F. & Monasterio, M. (Eds). High Altitude tropical Biogeography.Oxford University Press & American Museum of Natural History, New York : 387-403, figs.], Sienknecht [1992. Das Phänomen der Speziation in der Gattung Orestias aus dem Altiplano Südamerikas. Eine kritische Auseinandersetzung mit der Zuerkennung des Art-Status. Unveröff. Staatsexamensarbeit, Fach. Biol. Univ. Hamburg : 97 pp.], and Villwock [1993. Die Titicaca-See-Region auf dem Altiplano von Peru und Bolivien und die Folgen eingeführten Fische für Wildarten und ihren Lebensraum. Naturwissenschaften, 80 (1), Januar: 1-8, fig.] ; in-depth collecting of material for a DNA study by Incas tourist aquarists (at least) is eagerly expected …

* 33- The numerous names with missing types or undisclosed type material : this case is multiple and solving this case may be a prerequisite to further steps in research ; first, localisation of Aphanius types in Turkey (plus other countries) : despite the fact that Turkey is a country with stable Institutions and its Cyprinodont fauna is well known, type material of many species have not been located ; such as Aksiray's taxa (mostly synonyms) plus others (maybe elsewhere): aksaranus, altus, burdurensis, burduricus, flavianalis, fontinalis, iconii, litoralis, lykaoniensis, maeandricus, meridionalis, obrukensis, parvus, venustus, sureyanus, transgrediens, alexandri, boulengeri, mentoides, orontis, similis, splendens, or anatoliae (probably in Budapest), or lineatopunctatus, macrogaster, nanus, sarda, thermarum, timidus (all of them described from various parts of the Mediterranean basin); second, important valid species with missing types or undisclosed types : undisclosed holotypes of Fundulus chrysotus, catenatus, floridensis, Cyprinodon riverendi and Lucania affinis, no types "known" for Fundulus het. macrolepidotus, diaphanus diaphanus, dia. menona, majalis, luciae, notatus, Cyprinodon rhomboidalis, variegatus variegatus, var. riverendi, or Aphanius fasciatus ; third, no more types (they are reported to be lost in MNHN) for Valencia hispanica : a neotype from Cataluña should be designated ; fourth, no types disclosed for Fundulus badius, craticula, ornatus, swampinus, viridescens, multifaciatus, formosus, Fundulopanchax rubrofasciatus, Leptolebias fluminensis, L. sandrii, Leptolucania manni, Micropanchax keilhacki, Profundulus mexicanus, Orestias lastarriae ; this is a situation that is not acceptable and does not bear comparison with other groups of fishes ; let's end this list of challenges, mostly very difficult to solve, by a call to join our forces and soon clean in front of our door …

When these 33 challenges are resolved, a magnificent improvement will be achieved… but, sure, new challenges will arise !



III. Diagnostic challenges


The next challenges are certainly to build up more solid diagnoses or even to propose a first diagnosis for many valid Killifish species.

What is a diagnosis?

It is a definition, a statement that allows separation (from "dia", a Greek word) of knowledge (from "gnosis", a Greek word) ; with a diagnosis, a name at the species or at the genus level is defined as single, unique, in comparison to other, possibly related, names. With a diagnosis, the major issue is not to describe, but to separate.

Is it absolutely necessary ?

Yes, for 3 main reasons: first, a diagnosis is compulsory according to ICZN, the International Commission of Zoological Nomenclature (since 1930) ; second, a diagnosis is very useful and even unavoidable to build up knowledge on a given group (progress requires a step-by-step strategy), third, a diagnosis is today the major resource in building the systematic matrix of data that enables to develop phylogenetic trees with computer programmes.

Why are there diagnoses currently missing ?

The present unsatisfactory situation may originate from very old taxa (before 1930), from recently described taxa (by neglecting authors), from new data thatchange the composition of a group (new species of a known group, undisclosed data on live colour pattern or ecology (etc.) ; besides, DNA results havereshuffled several of our "certainties"), from the first livecollections of old taxa, etc.

Are diagnoses difficult to write ?

Yes and no. Obviously for a unique species, it is easy (Adamas formosus is the single Killifish "with a heart-shaped white blotch on front") for immediate characters (but other characters of that species, such as the faint dark vertical bar on eye or its larger eye or its unique behaviour, may not be easily disclosed) ; for a cryptic species, it may be very difficult, orsaid differently, it may need a detailed step-by-step analysis ; a typical diagnosis must list a number of characters (live pattern, preserved pattern, juvenile pattern, mood-driven pattern, bones, rays, micro-morphological features, etc.) that are shared with other names or alternatively that are not shared with comparative names ; and a character can be anything, such as the upper margin of the male Caudal fin, or the presence of a black spot in juveniles, or the shape of the Anal fin in adult female, etc. Today everybody has become familiar with this process, thanks to computer technology. Then, for a name, think of a simple character and answer yes (= "0") or no (="1") for it and the related names (or vice-versa) : this is the beginning of your diagnosis… not easy to start, but after the initial effort, it is just a brain exercise !

What are the critically missing comparative diagnoses for oviparous Cyprinodonts?

1. the Epiplatys fasciolatus-olbrechtsi superspecies, with more than 10 names and many that are difficult to set apart

2. the separation of Kryptolebias marmoratus and ocellatus, synonyms, subspecies or distinct species and their comparison with caudomarginatus and brasiliensis

3. the re-definition of all components of Fundulopanchax gardneri-mirabilis (more than 10 names)

4. the separation of all components of the Rivulus urophthalmus superspecies (more than 10 names)

5. the re-definition of the components of the Cyprinodon variegatus superspecies

6. the re-assignment of the Uruguayan populations of Austrolebias adloffi to Costa's new names

7. the comparative re-definition of the components of the Nothobranchius guentheri or korthausae superspecies

8. the comparative re-definition of the components of the Micropanchax loati/kingii superspecies

But this list is only an appetizer : actually most groups of Killifish wouldrequire a new diagnosis !

Who can help in securing and actually producing new diagnosis ?

Although the author's opinion may not be shared by all scientific schools, it appears that expert aquarists, notably those members of specialised study groups, have a lead on the issue. For several reasons : they own the live fish, they can observe their fish for long periods and at the various stages of their lives, they can compare related species by putting them close to each other (or, for males only, mixed together in a single aquarium), they can exchange their findings with other experts, they can help scientists in producing their matrix of data to improve knowledge. These expert aquarists are used to write articles on breeding and maintenance, with a fine and detailed description of their fish : for them, to add a diagnosis to their article will require a more rational approach (is that character of species "x" also found in species "y", "z", is that difference between "x", "y", "z", stable or not, etc.?)… After a first experience, it will be considered as very attractive and more useful than just a detailed description! Notably, if it is undertaken within a local group meeting.



IV. Aquaristic challenges


Cooperation from aquarists is also questioned from the interior, i.e. by the aquarists themselves : "what can I do to help?" is a very common question.

Standard contributions are obviously welcome, but for experts, aquaristic challenges are to be targeted in priority.

What are the requirements ?

The requirements are obviously scientific, not in terms of education but in terms of spirit.

* To be serious and curious, with modesty.

* To observe without any "a priori" and repeatedly (at least twice the same observation).

* To generate experimental observations, facts, and deductions without emotion.

Let's propose 4 candidate profiles, according to age, experience, owned devices and funds.

The first challenge concerns documentary. The first requirement to improve knowledge is obviously to know and acknowledge what others previously published. The issue is not to write a witnessing report, but to report on the current state of the art, then to separately bring new data by forwarding new observations, by discussing and criticizing old observations (by the way, present times push us to forget everything : who can scrutinize old aquarium literature to report on first breeders for Killifish species -most data are missing ? … It is a necessity -a "devoir"- of remembrance for those elderly aquarists who brought so much in the past).

The second challenge concerns collecting trips (if political conditions permit, obviously) : in this case, the issues are well known and are listed in Killi-Data books and online (where to go, when to go, the key success factors). With in addition, a kind request : to deliver findings (collecting localities, ecological observations and measurements) to others by publishing them and to keep material for researchers (only one specimen poured alive in 95% ethyl alcohol is enough to study genes and 2-3 specimens are minimum to study morphology and osteology… not a big burden). Sadly, many collecting trips have never been reported and all this knowledge is lost.

The third challenge concerns breeding : obviously the crossing experiments are never ended, most of the results have been produced by Scheel and collaborators, several decades ago (and since then, few reports) and for obscure reasons some people erroneously discourage amateurs by saying that crossings are not useful any more (by the way, how many crossings have been undertaken between recently described South American annuals? None !). If crossings sound outdated, then why not tackle the embryology issue ? It requires the purchase of a small microscope (priced today, less than 100 Euros or USDollars) and a strong curiosity : it is amazing how many details are available on the Killifish egg membrane, variably designed according to groups of species and how informative are the development steps of the embryo.

The fourth challenge concerns behaviour. Like for the today affordable microscope, this requires a lot of time and a new equipment that aquarists may already have purchased for their family : a simple digital video camera. Only 2 years ago aquarists could not imagine how they could bring value added on behaviour, but since then, the major contributions by the Brazilian Drausio Belote have been published [Boletim-do-Museu-Nacional-Rio-de-Janeiro-Zoologia 2002 28 Outubro; 489: 1-10, Reproductive behavior patterns in the Neotropical annual fish genus Simpsonichthys Carvalho, 1959 (Cyprinodontiformes, Rivulidae): Description and phylogenetic implications / Boletim do Museu Nacional, Nova Série, Rio De Janeiro - Brasil, Zoologia N° 514, 10 de Fevereiro 2004, 5 Figures, Reproductive Behavior Patterns in three Species of the south American annual Fish Genus Austrolebias Costa, 1998 (Cyprinodontiformes, Rivulidae) / Archivos Do Museu Nacional, Rio De Janeiro, 61 (4): 241-244, 6 Figures, 2003, Reproductive Behavior of The Brazilian annual Fish Cynolebias albipunctatus Costa & Brasil, 1991 (Teleostei, Cyprinodontiformes, Rivulidae): a new Report of sound Production in Fishes / Boletim do Museu Nacional, Nova Série, Rio De Janeiro - Brasil, Zoologia N° 515, 19 de Fevereiro 2004, 1-7, 4 Figures, Reproductive Behavior Patterns in the Brazilian annual Fish Plesiolebias glaucopterus (Costa & Lacerda, 1988) (Cyprinodontiformes, Rivulidae, Plesiolebiatina)]. These articles open a vast and promising new field of observations for all Killifish.



V. Conclusion

There are no minor contributions and no minor cooperation. Cooperation, according to each one's capacities, is open to all who want to know more, to understand more and to share their findings.

Future success is in front of us… even if complexity is very high with Killifish.

Hopefully these systematic, diagnostic and aquaristic challenges will be a boost to our community and a spur to speed up knowledge progress on Killifish !




Paris, August 2003-September 2004

Published in 3 parts in : Killi News (British Killifish Association), N° 469 (October 2004 : 104-108), N° 470 (November 2004 : 118-126), N° 471 (December 2004 : 131-138).

(first available online on March 9. 2004, as Killi-Data newsletter "Infoweb 8", then enlarged to the aquaristic challenges for the present purpose)


Additional reading of this author's publications ? Select any ONLINE article.