news:510AF258...@btinternet.com...
> Jack wrote:
>>
>> Do you think it *needs* proving that
>>
>> lim_(x --> oo)} (ln(x) / x) / (ln(x + ln(x)) - ln(x)) = 1
>>
>> (which is the central result in respect to the convergence I've been
>> asking
>> about) or can it be taken to be true without proof? After all, the
>> calculator says it's true....
>
> For a human proof of the above or the original you may wish to try
> L'Hopital's rule.
>
Thanks -- I got it in the end, after signing up for Wolfram Alpha Pro.