Determining butterfly weight distribution I'm a researcher studying a specific butterfly species. I've conducted an experiment where I measure the weight of this species at different locations on the earth. The data is discretized because it is recorded to the nearest microgram.
In some locations I have hundreds of measurements in other locations I have only a few measurements. How can I build a discrete distribution for overall butterfly weight from all of these locations?
My specific problem is that I'm having a difficulty with determining how the effect of variable sample size should be integrated into my calculations. How can this data be combined to generate an accurate distribution of butterfly weight for this species? 
Thanks for the help.
 A: For an overall density estimate, you can simply use boxplot or density plots, or any numeric summary you like (quartiles, means, medians etc). The locales with fewer specimens will naturally contribute less to any of these.  
If you would like to weight the locales differently than their 'natural' weight, then e.g. the density function in R lets you apply weights. Other methods of estimation will also let you apply weights. 
From what I can tell, applying these weights would make sense if your sample sizes were not proportional to the actual population of this species. You could then weight the samples to match the populations. 
A: It sounds like what you're asking about would lead one to simply combine all the data from each location into a single large sample and use that distribution as the single worldwide distribution of weights.
Imagine we only had three locations:A:   125  137  159  117  136  129  133  145

B:   144  137  119

C:   131  128

Then the combined sample of weights would be:
World: 125  137  159  117  136  129  133  145  144  137  119  131  128

