Hi David,

You might find the following useful.

Seetah et al, 2014. A geometric morphometric re-evaluation of the use of dental form to explore differences in horse (Equus caballus) populations and its potential zooarchaeological application. Journal of Archaeological Science, 41. 904-110.

Best wishes,

Alison

Alison Foster
PhD candidate
School of Archaeology and Ancient History
University of Leicester
University Road
Leicester
LE1 7RH

________________________________________
From: Analysis of animal remains from archaeological sites [[log in to unmask]] on behalf of Deb Bennett [[log in to unmask]]
Sent: 19 July 2014 07:49
To: [log in to unmask]
Subject: Re: [ZOOARCH] distinguishing horse, mule, and donkey

Ooh, me too, me too. I have resisted statistical morphometrics throughout
my professional life as grating on my sense of honesty as well as being
IMO a waste of time, but would eagerly learn any mathematical or
geometrical technique that really described shape. And I would think that
horse teeth would be a super-easy place to start. I once wrote a FORTRAN
(yucky yucky old dinosaur) algorithm applying Mandelbrot's chaos equation
by forcing a little ball to "walk around" the occlusal enamel pattern of a
horse tooth, in hopes that the "walkaround" would yield a unique and
thereby a diagnostic number .... it did, or seemed to, but Mandelbrot
himself told me "no it can't be unique" so I dropped the idea. Maybe
somebody now has a better way and if so I'd be eager to hear of it. Cheers
-- Deb Bennett

> Worth pointing out that while Deb is IMHO right that shape cannot be
> ascertained from linear caliper measurements, it CAN be both measured and
> statistically analysed using geometric (as opposed to traditional)
> morphometrics - GMM - and a landmark approach based on carefully taken
> photographs. This is often applied to occlusal surfaces of teeth,
> certainly for bears and pigs thus far.
>
> However, I don't know if this approach has yet been applied to horse
> teeth. Probably. If anyone does know of such work I'd be grateful for the
> reference. Otherwise, there's a project for someone!
>
> Best,
> David
>
>> On 19 Jul 2014, at 08:26, "Deb Bennett" <[log in to unmask]>
>> wrote:
>>
>> Jonathan, there is a large body of literature on this subject; the two
>> authors to go to first are Vera Eisenmann and Ann Forsten. For a handy
>> start on getting the references, look in the bibliography to Bruce
>> MacFadden's book "Fossil Horses."
>>
>> Though they take somewhat different approaches, both Eisenmann and
>> Forsten
>> are of the opinion that if you take enough measurements on enough
>> specimens, that you will then be able to tell equines apart
>> "statistically". I have never felt real comfortable with this, any more
>> than I feel that morphometrics per se can lead you to a grasp of the
>> true
>> relationships within any group of animals. In short, I think that the
>> thing that most powerfully and accurately distinguishes one sort of
>> animal
>> from another is shape, but shape cannot be ascertained by any set of
>> calipers, and cannot be specified by any manipulation, statistical,
>> algebraic, or otherwise, of the set of linearities that repeated caliper
>> measurements produce. In short: size cannot really proxy shape, no, not
>> even when you proxy it "to a limit of infinity".
>>
>> Nonetheless, using size to proxy shape is the current professional norm,
>> and unless you bow to the god of the caliper and the T-test, you are
>> more
>> likely to perish than be able to publish.
>>
>> So I'm not asking you to take my advice -- only giving it out, like just
>> any other old fogey -- my advice might help you solve the problem, it
>> might foster your level of insight as to what equines are all about, but
>> it's not likely to help get you a job. I think that the only RELIABLE
>> way
>> to tell equine species apart is to have complete, or nearly complete,
>> skulls. You can reference my "Stripes Do Not a Zebra Make" paper in
>> Systematic Zoology from 1982 to get the characters by which this
>> differentiation can be made.
>>
>> Shy of having complete skulls, complete jaw rami and/or complete
>> associated rows of cheekteeth are pretty good. One can usually tell a
>> horse's jaw from a mule's or donkey's on sight. If you've got teeth, you
>> can look at the depth to which the ectoflexid penetrates, the separation
>> of the metaconid-metastylid loops, and the shape of the linguaflexid. In
>> the upper teeth you can look at the shape and relative length of the
>> protocone and the degree of wrinkling exhibited by the enamel,
>> especially
>> that which rims the fossettes.
>>
>> Shy of skulls, jaws, or teeth, you have cannon bones -- those of asses
>> are
>> smaller and more slender, those of horses usually larger, but even if
>> not
>> larger (or longer), then certainly broader and stouter, especially the
>> fore cannons. Mules will be intermediate. You can also pretty well make
>> the same assessment on proximal phalanges, i.e. horsemens' term "long
>> pastern bones", or indeed on any of the phalanges, again especially
>> those
>> pertaining to the forefeet.
>>
>> Next best after that would be just about any limb bone, the best perhaps
>> being the pelvis. If very small, it's almost certainly a donkey; if very
>> large, and especially if the ischium is relatively long, it's almost
>> certainly a horse.
>>
>> Thoracic vertebrae esp. about T4-T8 of horses have longer dorsal spines
>> than those of donkeys, because domestic horses have been bred to have
>> high
>> withers, whereas donkeys almost always retain the primitive condition,
>> which was pretty muttony.
>>
>> Can't help you too much beyond this, except to say that it is very
>> unwise
>> to stick your neck out farther than a tentative assignment or "c.f." for
>> anything shy of a complete or nearly complete skull. If you have a lot
>> of
>> loose unassociated teeth, you may very well be better off with Eisenmann
>> and Forsten than me.
>>
>> For the other old fogeys reading here, by the way, I had the amusing
>> experience this evening (at a wedding reception) of teaching a teenaged
>> boy how to use a manual typewriter (we were to type messages of
>> congratulations to the happy couple, who are also both silverhairs).
>> This
>> young man now knows what a "ribbon reverse" and a "warped platen" are.
>> What will the world come to next. Cheers -- Deb Bennett
>>
>>
>>> Hello All,
>>>
>>> I am working on a faunal corpus containing equid bones. For the period
>>> under question there were known to be horses, donkeys, and mules at the
>>> site. How can these be distinguished (esp. mule and horse)? Any tips,
>>> references, or helpful advice will be greatly appreciated.
>>>
>>> Best,
>>> Jonathan
>>>
>
>
>