Forewing length vs. wingspan

[Chuck]

I experience a real shock today, totally blew me away. This concerns visual appearance of size vs. an actual measurement.

Historically in the olden days I'd measure wingspan for discussion & comparison. But the research publications I've been reading have been using Forewing Chord Length (FCL) which is a straight line from base to apex. Plus, FCL is more accurate, since wingspan is dependent upon how the specimen is set (let's face it, I'm not Trehopr, to paraphrase John from memory "I stick #3 pins in the wings and set it.") So I figured I'd use FCL to determine FCL range in a series of the same taxon.

I picked a few average females and measured them. The measured FCL 52-54mm. Then I picked a really big, round fat female: 52mm. Huh? I measured a bigger female: 53mm. I measured a small female: 54mm. Huh? They're almost all 52-54mm no matter how broad looking the wings appear, how fat the body is.

So I figured I'd measure males, which of course are always smaller. Same results: FCL consistently 52-54mm. This is rediculous; I can look a series of 18 males, all captured same place, same day and say "this is a big one, this is a small one." But nope, I had to go back ten minutes later and measure again. Same results.

One take-away is that visual appearance of large vs. small is misleading. Some specimens surely had a much longer FCL...but they did not.

So, OK, what did FCL tell me? That the measurement is quite consistent (yes, there are a very few outliers) even between sexes. That's nice, but what does it mean? On that series of 18 males the very smallest FCL was 48mm and largest 53mm; spans were 70mm and 83mm. It almost seems like span tells me more about the size.

[note on above, the 90 degree triangle guys: triangle (not actual wing) area is: smallest 575, largest 697] Still though, I can take others that are the same FCL and due to wing shape variation (trailing edge of wing) they will be a bit different- I can see the overall difference.

Does the scientific value of FCL depend on the smallest angle (remember Pythagoras)? Take for example, a series of Papilio antimachus with those long wings, where span is "roughly" FCL*2+body. Contrast that with a Graphium, where span is nowhere near FCL*2+body.

So what is the value of FCL? Yes, it can be consistently measured. But it doesn't really tell us about the actual size of the specimen, like wingspan can. I know, it's "one data point" you'll say. But why did so many publications get away from wingspan?

[adamcotton]

So what is the value of FCL? Yes, it can be consistently measured. But it doesn't really tell us about the actual size of the specimen

Forewing length can be very useful when comparing between different taxa. There may also be differences between generations of the same taxon.

It is a good idea to quote range (e.g. 48-54mm) and average (e.g. 52.5, n = 18) where n is the number of specimens measured.

But why did so many publications get away from wingspan?

As you indicated, wingspan depends on the way a specimen has been spread, and also is affected by the width of the thorax. Another advantage of using forewing length is that it can be measured just as easily from papered (unprepared) specimens. So if you have a large number of samples that have not been spread you don't need to spend a very long time spreading all of them before obtaining measurements.

Adam.

[Chuck]

Thank you, Adam.

Another question: what about outliers? If 95% of specimens are 48-54mm, but a few are 42mm, it seems misleading to say 42mm-54mm. Yes, that is indeed the range, but not the norm.

[adamcotton]

That's why the average and number of samples helps. If most are in the middle of the range the average will also be approximately in the middle too, assuming a reasonable number of samples. If the average is skewed to one side and the 'n' value is low, then it is obvious that an outlier has affected the average. If there are occasional extreme odd specimens it may be worth mentioning them in the text.

Adam.