[R] Illustrating kernel distribution in wheat ears

[Carl-Göran CG. Pettersson]

Thanks a lot for the quick response!

The suggested code worked fine up to a certain point: the actual plotting from the datasets...
With the sample dataset in "samp" my code looks like this at the end:

> for(cultivar in 1:6)
+   boxplot(unlist(samp[samp$cn==cultivar[cultivar],]),
+    add=TRUE,at=cultivar)
There were 12 warnings (use warnings() to see them)
> warnings()
Warning messages:
1: In min(x) : no non-missing arguments to min; returning Inf
2: In max(x) : no non-missing arguments to max; returning -Inf
.
.
.
.
.

What is happening and why?
/CG
________________________________________
Från: Jim Lemon [jim at bitwrit.com.au]
Skickat: den 11 januari 2010 10:56
Till: Carl-Göran CG. Pettersson
Kopia: r-help at r-project.org
Ämne: Re: [R] Illustrating kernel distribution in wheat ears

On 01/11/2010 06:47 PM, Carl-Göran CG. Pettersson wrote:
> Dear all
>
> R2.10  WinXP
>
> I have a dataset dealing with the way different wheat cultivars build their yield.
> Wheat ears are organised in spikelets where the spikelets can be numbered from the bottom, with even numbers on one side and odd on the other.
> I know how many kernels there were in each spikelet after some months spent counting them...
>
> Now I want to illustrate the differences between the cultivars in how the kernels are distributed in the ears.
> In the best of all possible worlds it would be possible to place histograms or boxplots on adjecent sides of vertical lines representing different cultivars.
> I have done some experimenting using boxplot() but I am stuck and out of ideas right now.
>
>
Hi Carl,
Is this what you are looking for?

plot(0,xlim=c(0.5,6.5),ylim=c(0,6),
  main="Kernel distribution",xlab="Cultivar",
  ylab="Count",type="n",xaxt="n")
cultivars<-unique(spikernel$cn)
axis(1,at=1:6,labels=cultivars)
for(cultivar in 1:6)
  boxplot(unlist(spikernel[spikernel$cn==cultivars[cultivar],]),
   add=TRUE,at=cultivar)
abline("v"=1.5:5.5)

Jim