>> Greetings.  I happened to notice that in evaluating R source-code blocks in
>> an Org-Mode file I got some cruft from my .Rprofile in the output on some
>> occasions but not on others.
>>
>> To my surprise, it appears that the CASE of the keywords on the BEGIN_SRC
>> line affects the output.  Please see the appended for a simple example.
>>
>> I haven't been able to find this behavior documented any place, but of
>> course it's possible I just missed it.
>>
>> Can somebody confirm/deny that Org-Mode is SUPPOSED to behave this way?
>>
>> Thanks,
>>
>> -- Mike
>>
>>
>> * Mike's test of UPPER CASE keywords
>>
>> #+BEGIN_SRC R :RESULTS output :EXPORTS both
>>
>>     x <- rnorm(5)
>>
>> #+END_SRC
>>
>> #+results:
>> |   -1.78820283900565 |
>> |  -0.469227269149123 |
>> |   -0.92353467915834 |
>> | -0.0831621941877021 |
>> |    -1.5583681234565 |
>>
>>
>>
>> * Mike's test of lower case keywords
>>
>> #+BEGIN_SRC R :results output :exports both
>>
>>     x <- rnorm(5)
>>
>> #+END_SRC
>>
>> #+results:
>> :
>> :
>> : All ready to go...
>> :
>> : Fri Aug 19 15:46:15 2011
>> : Hasta la vista!
>
> Hi Mike,

> I believe what is happening is that :RESULTS and :EXPORTS are not
> recognized, so :results (the correct header argument) defaults to value.
> 
> hth,
> Tom
> 
> -- 
> Thomas S. Dye
> http://www.tsdye.com

Thanks, Tom.  I think your explanation is correct.  The manual page:

    http://orgmode.org/manual/Specific-header-arguments.html#Specific-header-arguments

does indeed show all of the arguments in lower case.  I had seen BEGIN_SRC and
begin_src used interchangeably and so assumed that same latitude applied to
the keywords.  And I have a slight preference for putting keywords in upper
case and values in lower case, just to distinguish one class of things from
the other.  My bad.

-- Mike