It occurred to me this morning that I should probably run
a test of the two means, just to put a statistical cap on
the measurements I sent out yesterday. I made all measurements
using a vernier caliper to the nearest 0.1mm then computed the
ratios as per Phil's instruction. I also added Denny5
(h/b = 0.48) giving me a total of n = 6 for the Renton gull
after reading Phil's follow-up.

Using: http://www.graphpad.com/quickcalcs/ttest1.cfm?Format=C
because it's a whole lot faster than doing it by hand, I obtained
the following:

Unpaired t test results

P value and statistical significance:
The two-tailed P value equals 0.3152
By conventional criteria, this difference is considered to
be not statistically significant.

Confidence interval:
The mean of Group One minus Group Two equals 0.0250
95% confidence interval of this difference: From -0.0257 to 0.0757

Intermediate values used in calculations:
t = 1.0317
df = 19
standard error of difference = 0.024

Mike Patterson calculated:
>
> E&T = Enticott and Tipling
> Grant = Grant
> O&L = Olsen and Larsen
>
> E&T 8 0.44 Sullivan1 0.54
> E&T 9 0.52 Sullivan4 0.48
> Grant 294 0.48 Denny1 0.48
> Grant 297 0.47 Denny3 0.50
> Grant 301 0.45 Denny6 0.45
> Grant 302 0.48
> Grant 305 0.54
> Grant 308 0.59
> Grant 309 0.44
> O&L 129 0.49
> O&L 134 0.58
> O&L 137L 0.60
> O&L 137R 0.58
> O&L 138 0.50
> O&L 140 0.54
>
> Average 0.51 0.49
> st dev 0.05 0.03
>


--
Mike Patterson
Astoria, OR
celata at pacifier.com

Half-a-bee, philosophically must ipso-facto half not-be.
But half the bee, has got to bee Vis-a-vis its entity...
d'you see?
But can a bee be said to be or not to be an entire bee
When half the bee is not a bee due to some ancient injury?
-Monty Python

http://www.pacifier.com/~mpatters/bird/bird.html