Contributed by MelT

Or should that be a C. frutescens tepin?

The C. annuum/C. frutescens/C. chinense complex is a long standing scientific puzzle, especially the question of the wild precursors of C. frutescens and C. chinense.  From what I've read, it seems unclear whether truly wild [that is, never cultivated] forms of C. frutescens and C. chinense are known to exist.  It is also unclear whether these two groups are "really" separate species that would have had different wild ancestors, or whether they would have come from a common wild source.   

Researchers have repeatedly suggested that C. chinense and C. frutescens should be combined, that together they are a single complex of related cultivated and semi-wild forms. Some say they should be included with C. annuum as a single species. If the two are combined as one species [but kept separate from C. annuum] the name "Capsicum frutescens L." has  priority. 

[It makes sense to me that truly wild C. chinense/frutescens populations might now be rare or even extinct, at least in their original form.  Any wild populations growing near related cultivated chilis could have a  problem from producing weedy hybrids with the crop forms; they would likely be competing with and interbreeding with the continually escaping hybrids, and might soon be genetically swamped by their  mixed descendants.]   

An obviously related question concerns just what we would expect to see in truly wild C. chinense or C. frutescens [or in a single wild ancestor of both domesticated groups].  Based on the existing primitive varieties of C. chinense and C. frutescens and comparisons with the wild tepins belonging to C. annuum v. glabriusculum,  they would be:

-- shrubby plants  with small, red, round, pungent, deciduous, tepin-type fruits on erect stalks, [traits shared with wild C. annuum tepins]
-- with shinier leaves, greenish flowers in pairs or small clusters [distinctive shared traits of C. frutescens and C. chinense.]

The "Aurochs" approach
Whether or not the ancestral wild forms still exist in nature, it may be possible to make hybrids among different relatively primitive cultivated chili varieties to produce something very close in form to their wild ancestors.  ["Primitive" simply means an organism retains features present in an ancestor that have been lost or changed in other descendants of that ancestor.  There are a number of known C. chinense and C. frutescens varieties or accessions that are comparatively little changed from wild tepins.] 

This idea is a bit reminiscent of the early 20th century breeders who attempted to re-create the aurochs, the extinct wild ancestor of domesticated cattle, by crossing different primitive cattle strains.   The scientific value of the results are disputed [and the whole project was tainted with  a Nazi political connection], but they did get an aurochs-like breed of "Heck Cattle".  There reportedly are current plans to use modern genetic methods to develop strains of cattle that are a closer genetic match to the extinct form.

http://en.wikipedia.org/wiki/Aurochs
http://en.wikipedia.org/wiki/Heck_cattle

Wild-type, primitive traits commonly behave as genetic dominants.  For example, hybrids between yellow fruited chili varieties and ones with red fruits have uniformly red fruits in the F1 generation.  Hybrids between different "primitive" chili varieties might be expected to immediately express a new combination of the primitive traits from both parents, and appear closer than either parent to a "wild chili".

One cross I made  in 2009/2010 seems to have accomplished just that fairly dramatically,  in a single step. 

I'd like to stress that this wasn't really any big project involving lots of time, space, and effort. Or much real knowledge, careful planning, or adequate research beforehand, for that matter. They're just a small number of hybrids that were made among a few varieties in my little chile garden.  I'm aware that there are various other tepin-like "wild C. chinense" varieties and accessions already being grown, some of which may already be about as primitive as this new hybrid. 

The parents of this hybrid were both obtained from the same commercial source:

1] C. chinense "Wild Brazil"

Apparently the same or similar varieties are also called "Cumari do Para", or "Brazilian Cumari" and other names.  The plants appear to be dwarfed overall, low growing with small leaves, and have unusually tiny flowers even for a wild chili.  It has tepin-like yellow fruit with a strong C. chinense aroma.  Flowers and fruits are erect, more primitive in this than most other C. chinense with their pendent flowers and fruit.

2] A plant I bought as "Wiri Wiri" that is much more problematic.

Although it was sold as C. chinense, I think this variety may better be considered C. frutescens.  It is clearly different from the C. chinense varieties also being grown as "Wiri Wiri"  [more on this separate problem in another post]. For now, I'll continue to call it "Wiri Wiri" since I have no other name for it.

I only had a few suitably "primitive" C. chinense/frutescens chili varieties to use. Besides these two, a few crosses were also made using "Duke Pequin"
http://wildchilli.eu/index.php/cfrutescens-duke-pequin

and "Mariwiri"
http://wildchilli.eu/index.php/qmariwiriq-from-guyana

Although it shouldn't be surprising that a hybrid between two tepin-like varieties would itself  be tepin-like, I still think the results were impressive.  The "Wiri Wiri" X "Wild Brazil" F1 plants are somewhat taller-growing than either parent:

They are somewhat smaller-fruited than both parents, and the round red tepin-type fruits are more easily separated from the calyx when ripe.

The tiny red tepin fruits are erect on long slender stalks:

They were more attractive to birds than either parent. 

The above image shows several empty calyx cups; those fruits mostly having been taken by birds soon after they turned red.

Fruit comparison, the hybrid between the two parental forms (the single red fruit above the two "wild brazil" on the right is a wild C. annuum v. glabriusculum "chiltepin, Texas" to show its similarity):

The F1 hybrid is similar enough in overall appearance to a wild C. annuum v. glabriusculum tepin that it might be difficult to pick out the plants from among a group of C. annuum tepins. The hybrid has glossier leaves, greenish flowers in pairs, a differently-shaped fruiting calyx, and its crushed fruit have a C. chinense aroma.

It's hard to believe this "Wiri Wiri" X "Wild Brazil" hybrid isn't simply a wild tepin:

Another relevant hybrid was "Duke Pequin" X "Wild Brazil". The F1 were more robust plants than the previous hybrid or their "Wild Brazil"  parent:

[plant partly defoliated by excessive heat and frequent water stress. It had been a lot prettier.]

They were very productive of slightly-elongate fruits.

Fruit comparison:

Later in the season the plants were covered with bright red fruits [despite the birds' continuing efforts]

Fruits of both of the above crosses have a noticeable  "C. chinense" aroma and flavor lacking in the two C. frutescens-like parents.  Like other tepin-sized chilis, they dry quickly without special treatment. 

A further cross, "Duke Pequin" X "Wiri Wiri" generally seemed close to their "Wiri Wiri" parent, but with some elongation of the tip of the fruit.  The reciprocal cross was closely similar.

Full-sized unripe fruit:

Comparison ["duke pequin" on the left, "wiri wiri" on the right, hybrid between]:

Some flower comparisons:

Upper row from left: "Wild Brazil"; "Wiri Wiri"; "Duke Pequin" [note the tiny size of "Wild Brazil" flowers]
Lower row: "Wiri Wiri"X"Wild Brazil"; "Duke Pequin"X"Wild Brazil"

Further fruit comparisons:

Top row, from left: "Duke Pequin";  "Wiri Wiri"; "Wild Brazil"; "Mariwiri"
Middle row: "Duke Pequin"X"Wiri Wiri"; "Wiri Wiri"X"Wild Brazil"
Bottom row  "Duke Pequin"X"Wild Brazil"; "Mariwiri"X"Duke Pequin"

F2 plants of all of these crosses should vary in detail, but generally should stay similar to their parental strains at least in their shared characteristics.  I expect yellow-fruited forms will reappear in 1/4 of the F2 generation of crosses involving "Wild Brazil". 

Further crosses among these hybrids, and adding additional primitive chili varieties to the mix would be interesting.  Others may want to try similar hybrids using other very primitive C. chinense varieties such as:

charapita
http://gardener.wikia.com/wiki/File:Chilli_Charapita.jpg
http://webzoom.freewebs.com/ruudendiana/Database%20Peper/Peper%20Charapita.jpg
http://www.pbase.com/chiles400/image/33354459
Yellow fire PI 260501
Grif 9239-AB; 
http://www.pbase.com/chiles400/usda
Yellow bird
http://petterssononline.com/habanero/peppers.php?action=variety&id=153
http://www.pbase.com/chiles400/image/24150077
tettinas de monk
http://www.thechileman.org/results.php?page=10&chile=1&find=Any&heat=Any&origin=Peru&genus=Chinense
PI 260504
http://www.thechileman.org/results.php?find=260504&heat=Any&origin=Any&genus=Any&chile=1&submit=Search
Capezzoli di Scimmia
http://www.pbase.com/chiles400/c

I'm not sure any of this necessarily says much about the species status of C. chinense and  C. frutescens.  Still, as others have noted, the most primitive varieties of C. chinense and  C. frutescens seem a lot closer together than the more familiar cultivated forms, and if the latter didn't exist there would be little reason to separate the "two species".