GarryJolleyRogers - Wed Nov 25 2009 - Version 1.5
Parent topic: BestPractices

UnnamedBodyPartsDefinedByLocation

The following descriptions are from Wheeler, Ants of the Congo. I have added sentence numbers.

• Subgenus Myrmocladœcus Wheeler ( =Myrmorhachis Emery, 1920; not of Forel, 1912)
  • 1. Pronotum very broad, with a lateral, lamelliform margin, often vaulted.
  • 2. Thorax rapidly narrowing behind.
    
  • 3. Epinotum very narrow at its sloping face, which often has a peculiar appendage.
  • 4. Gaster broad, short, and small, sometimes more or less spherical.

• Subgenus Manniella Wheeler
  
  • 5. In the maxima worker the anterior truncated portion of the head is strongly carinate at the sides and posteriorly depressed; the front is strongly depressed between the carinæ, the depression margined behind with an elevated ridge.
  • 6. The remainder much as in Myrmeurynota.

• Subgenus Myrmomalis Forel
  • 7. The entire body depressed in the worker and female, especially in the worker of C. obtritus which is completely flattened.
  • 8. Head rectangular in the worker major; elongate, trapezoidal in the worker minor; eyes placed laterally and behind the middle.
  • 9. Dorsum of the thorax flat; scale of the petiole low and thick.
  • 10. Integument black, opaque and pilose.
  • 11. Legs long, compressed, hirsute.

From these sentences I deduce the following body parts hierarchy. Each deduction is supported by the sentence number in parentheses at end. I only list parts that are mentioned in these treatments. Obviously there may be other parts. Some of this hierarchy might provoke disinterest from the myrmecologists. The real question is whether these should all be in a parts hierarchy—I doubt it—or just become special characters, e.g. “anterior truncated portion of the head depressed” as a character. But this seems special also.

I wonder if the real problem is that we only allow states to be modified, not characters….

I number the assertions below for ease of discussion.

• A body has a
  • A. gaster,
  • B. pronotum,
  • C. head,
  • D. thorax, and
  • E. epinotum.

• B. The pronotum has
  • B1. a lateral margin (1)
• C. The head has
  • C1. an anterior portion (5) which has
    • C11. sides (5)
    • C12 a posterior (5) [or is “posteriorly depressed” a character of the anterior of the head?]
  • C2. eyes (8)
  • C3. a middle (8) [“eyes placed behind the middle”]
• D. The thorax has a
  • D1. behind part (2)
  • D2. dorsum (9)
• E. The epinotum has
  • E1. a face (3) [maybe an epinotum has many faces, one of which is sloping??]

Surely many of these don’t belong in the parts hierarchy but in some general hierarchy of location terms. Probably these are exactly what specialists would say “yes but” to. Since I haven’t tried to code the descriptions yet, not sure what the impact is.

-- Main.BobMorris - 12 Jul 2004

Some of this may become clearer (not that the problem would go away, it just may get clearer) when we switch to a car. A car has a front (anterior), a rear (posterior), a top and bottom (= dorsum) side, it has left and right sides (= lateral). Some parts of a car may have been given names, like windshield for the front glassy part. However, orientation and part naming may be used interchangeable: front window or windshield. And the front has sides and a middle. Is the middle of the front the same kind of part category as a wheel is a part?

I completely agree that this is confusing. Dorsum = dorsal region of body. But is this a problem? I see the problem much rather in that statements may refer to combinations of parts and to subparts separately. But Here I see no big difference whether I say the entire plant is hairy or whether I say leaf glabrous, but hairy at the tip.

By the way: Is really the entire plant hairy? Probably not, just like in a black car the windows are probably still translucent...

Re: "I wonder if the real problem is that we only allow states to be modified, not characters." How could you modify a character? Can you give an example? A statement in my thinking implies a character (= variable) and value. I can only modify a statement, not the character. But I may be wrong. Prometheus II is seeing things slightly different as well.

Speaking of Prometheus: They define "Regions", so parts have generic regions that are applicable to all parts. In SDD these would be spatial modifiers, but they are expressable only in combination with statements, not parts itself.

-- Gregor Hagedorn, - 12 July 2004