Bob,
Interesting calculations. I have one comment. I don't know what the
distribution of the number of teeth looks like, but I am assuming that it is
normal. I also don't know how large your standard deviations are. But recall
that a normal distribution is symmetric by definition. Thus, you don't
double the sample number to calculate the chances of finding 2 specimens an
equal standard deviation units away on each side of the mean. For example,
if in helvola the hypothetical specimens with 10 or 21 labial teeth are just
above the observed range, I will assume that either specimen is about 4
standard deviations away from the mean. Since the distribution is symmetric,
on the average 0.13% of the  specimens will be in the range 3-4 standard
deviations smaller & another 0.13% of the specimens will be in the range 3-4
standard deviations larger than the mean. In other words, in a 1000-specimen
sample, one may expect to have about 1 shell with 10 & 1 shell with 21
teeth. The actual percentages depend on the actual standard deviations.

Of course, shells with 10 or 21 teeth don't have to exist just because
statistics predict that they would. There may be genetic constraints
preventing extreme shells from developing.

Aydin

>-----Original Message-----
>From: Bob Dayle [mailto:[log in to unmask]]
>Sent: Wednesday, May 01, 2002 6:43 PM
>To: [log in to unmask]
>Subject: Some labial tooth statistics of some cowries...
>
>
>For those _really_ interested in cowries, I have explored the
>range of possible labial tooth counts for a few of the Indo-
>Pacific species of Cypraea with breeding populations in Hawaii.
>
>By using the average labial tooth count and its standard deviation
>from each of these four species, I have "inflated" their populations
>to determine how big it would need to find a cowry with a labial
>tooth count greater or smaller than the current range of labial teeth
>in the samples that I am working with. A small table of possibilities
>looks like this:
>
>             my range of      # shells in   ("new" popul.)  to get 1
>Cypraea       lab. teeth       my sample     # of shells    shell w/
>
>helvola        11-20         2,600+         10,000       10 or 21 th
>fimbriata      17-26           500           1,150       16 labial th
>caputserpentis 13-22           300           5,000       12 labial th
>poraria        15-23            90             450       14 labial th
>
>So, to have a definite shot at finding a _Cypraea helvola_ with
>either 10 or 21 labial teeth, the chances are about one in ten
>thousand. (Yes, just like a lottery, or something.) To get one
>of 10 labial teeth AND one of 21 labial teeth, the sample would
>need to be nearly 20,000 shells! (That is, in Hawaii; populations
>in your area may vary.)
>
>Perhaps others have stats from cowries in their locales...
>
>Aloha,
>
>makuabob (a.k.a. Bob Dayle)
>
>_________________________________________________________________
>Join the world's largest e-mail service with MSN Hotmail.
>http://www.hotmail.com
>