Taxonomy & naming
Characodon garmani was described by David Starr Jordan and Barton Warren Evermann in 1898, in their monumental Fishes of North and Middle America (Bulletin of the United States National Museum). The species belongs to the order Cyprinodontiformes, family Goodeidae — the splitfins — and within it the subfamily Goodeinae, the live-bearing goodeids of the Mexican highlands and adjacent basins. Eschmeyer's Catalog of Fishes treats Characodon garmani Jordan & Evermann, 1898 as a valid name, and FishBase carries it under SpecCode 6232.
The specific epithet honours Samuel Garman (1843–1927), the American ichthyologist and herpetologist who in 1895 had examined the same material and assigned it to Characodon lateralis; Jordan and Evermann subsequently separated it as a distinct species and named it for him. The genus Characodon was erected by Albert Günther in 1866, its name combining Greek charax (a pointed stake or palisade) with odous (tooth) — "a tooth like a pointed stake," a reference to the dentition.
The taxonomic status of C. garmani is genuinely uncertain. The species is known from only a single specimen, and Characodon lateralis Garman, 1895 stands in part as a synonym of the earlier identification. Circumstantial evidence assembled by later workers (Artigas-Azas, unpublished) suggests the type may actually have come from a spring near Durango rather than from Parras, in which case it would represent one of the surviving Characodon (C. lateralis or C. audax) rather than a separate taxon. Pending resolution, it is carried as a valid extinct species.
Morphology
Everything known of the appearance of Characodon garmani derives from one preserved animal: the holotype, a mature female of 1 in standard length, held at Harvard's Museum of Comparative Zoology (MCZ-27704) and collected by the botanist Edward Palmer. No male was ever taken, and no living fish was ever described, so the species' size range, fin form, and the structure of any male intromittent organ are unknown.
The colouration recorded by Jordan and Evermann from that single female describes a fish olive to reddish-brown in ground colour, with scattered small darker spots on the back, a darker band running along the flank — with or without dark spots within it, and more distinct toward the rear of the body — and fins finely peppered with dark dots. This is a modest, cryptic pattern broadly consistent with its congeners, the living Characodon of Durango.
Because the description rests on so little, the morphology section can be said no more firmly than this: data are sparse. As a goodeid in the genus Characodon it would have been a small, deep-bodied splitfin in the size class of C. lateralis and C. audax, but every quantitative or sex-specific detail beyond the 1 in female holotype is simply unrecorded.
Habitat
Characodon garmani is associated with the springs near Parras de la Fuente in Coahuila, north-eastern Mexico — an endorheic desert valley within the Laguna de Mayrán subbasin, part of the internally drained Río Nazas–Aguanaval system (the lagunas de Mayrán y Viesco). These were isolated spring-fed waters typical of the closed desert basins of the Mexican Altiplano, where small fishes persist in springs, spring-runs, and short endorheic streams cut off from any outflow to the sea.
The type locality is given as "Parras, Coahuila, Mexico," though that attribution has long been doubted; the alternative hypothesis places the original specimen at a spring near Durango. Either way, the habitat was a desert spring system — clear, thermally stable groundwater-fed water of the kind that supports many of Mexico's most narrowly endemic and most threatened fishes.
The critical fact about this habitat is that it was altered early. The springs and associated waters around Parras were modified by the end of the nineteenth century and stocked with exotic fishes before the native fauna could be properly surveyed. Whatever the Parras characodon required, those conditions had largely disappeared within a generation of its collection.
Feeding
No direct observations of the diet of Characodon garmani exist — the species was never studied alive, and its feeding ecology is unrecorded. FishBase offers only a modelled estimate, placing its trophic level at about 2.2 (±0.1), which is consistent with a low-level omnivore rather than a predator.
Inference must therefore come from its living relatives. The surviving Characodon of Durango, C. lateralis and C. audax, are small omnivorous splitfins that graze on algae, aufwuchs and detritus and take small aquatic invertebrates and insect larvae from the substrate and water column of their springs. A comparable diet is the most reasonable expectation for the Parras characodon.
Beyond that analogy, the record is empty. There are no gut-content studies, no field feeding observations, and no aquarium notes, because the fish was extinct long before live specimens or modern study were possible. Data here are sparse by necessity.
Mating
Characodon garmani was a goodeid, and the goodeids of the subfamily Goodeinae are viviparous (live-bearing) fishes with internal fertilisation — so its reproduction was certainly internal. No mating behaviour was ever observed, however, and the species is known only from a single female, so nothing specific can be said about its courtship or the male anatomy involved.
It is important to note that goodeids differ from the more familiar New World livebearers of the family Poeciliidae (guppies, mollies, swordtails). Poeciliid males fertilise females using a gonopodium — a rod-like intromittent organ formed from modified anal-fin rays — whereas goodeid males do not possess a true gonopodium. Instead, the anterior rays of the male's anal fin are shortened and partly separated to form a structure called the spermatopodium (the andropodium), used to transfer sperm. The Parras characodon would have followed this goodeid pattern rather than the poeciliid one.
Goodeids also do not store sperm for the months-long, repeated broods seen in poeciliids, and superfetation — the carrying of overlapping broods at different developmental stages, well known in some poeciliids — is not the goodeid mode. For C. garmani specifically, none of this was observed; it is reconstructed entirely from family-level biology, and the species' own mating system is unknown.
Breeding
As a goodeid, Characodon garmani would have been a true livebearer that gave birth to well-developed free-swimming young rather than laying eggs. The defining reproductive feature of the family is matrotrophy: goodeid embryos are nourished by the mother throughout gestation through trophotaeniae — ribbon- or rosette-like extensions of the hindgut that emerge from the vent of the developing young and absorb nutrients from the maternal ovarian fluid, functioning somewhat like a placental connection. This is a far more intimate maternal investment than the largely yolk-fed development of poeciliid livebearers.
No brood of C. garmani was ever seen, so its gestation length, brood size and fry size are unknown. By analogy with its small congeners C. lateralis and C. audax, broods would have been modest — on the order of a few to perhaps a couple of dozen relatively large, advanced young per cycle — but this is inference, not record.
Goodeids generally do not practise superfetation and do not produce the rapid succession of broods from stored sperm that makes poeciliids so prolific; their reproduction is slower and the young fewer but better provisioned. For the Parras characodon, the breeding section ultimately reduces to a single certainty — it was a matrotrophic livebearer — wrapped around an absence of any direct data.
In the aquarium
Characodon garmani has no aquarium history. It was never kept, never bred in captivity, and no captive population of any kind exists or ever existed; it was extinct before the modern hobby of maintaining Mexican goodeids developed. There is nothing to record here in the way of husbandry experience.
What aquarists can do is keep its living relatives. The Goodeid Working Group and allied conservation-aquaculture efforts maintain ark populations of many imperilled Mexican goodeids — including the surviving Characodon species — precisely because so many wild populations are gone or going. These fishes are cool-to-warm spring dwellers that need clean, well-oxygenated, hard alkaline water, generous water changes, and a varied omnivorous diet, and they breed readily in dedicated tanks.
For the Parras characodon itself, the only honest entry is its absence: it stands as a reminder of why the captive maintenance of related goodeids matters. The fish that could have been kept is gone, and the hobby's role with this genus is now part of the conservation effort rather than ordinary fishkeeping.
Conservation
Characodon garmani is assessed as Extinct (EX) on the IUCN Red List (species ID 4530, assessment 3000349; assessed 18 April 2018). It is an endemic of the Parras Valley region of Coahuila whose habitat was destroyed or modified roughly between the late nineteenth century and the mid-twentieth, with no confirmed records in over a hundred years. Lyons and colleagues (2019), in their review of the distribution and conservation of Mexican Goodeidae, likewise classed it as extinct with no records since the 1890s, recognising it as a single evolutionarily significant unit with no surviving or captive populations. Under Mexico's NOM-059-SEMARNAT-2010 it carries no risk category — a reflection of its presumed loss rather than its security.
The cause was the early and rapid alteration of its desert-spring habitat. The waters around Parras were modified and stocked with exotic species before the end of the nineteenth century, and surveys from the 1940s onward have failed to recover the fish. This sequence — spring modification, water extraction, and the introduction of non-native fishes — is the same one that has driven much of the genus Characodon and many other Mexican Altiplano fishes toward the edge.
The one ambiguity that softens the record is taxonomic, not hopeful: if the type specimen in fact originated near Durango, C. garmani would prove to be a surviving Characodon under a different name rather than a lost species. Until that question is resolved, it is carried as a valid, extinct taxon — and its story remains a clear warning about how completely an endemic desert-spring fish can disappear before science has even met it alive.