This is from Stewarts's Calculus Early Transcendentals 8e, Chapter 4, Problem Plus, #8. The topics discussed in chapter 4 are: "Maximum and Minimum Values", "The Mean Value Theorem", "How Derivatives Affect the Shape of a Graph", "Indeterminate Forms and L'Hospital's Rule", "Summary of Curve Sketching", "Graphing with Calculus and Calculators", "Optimization Problems", "Newton's Method", and "Antiderivatives".
The problem is: $$\lim_{x\to\infty}\frac{(x+2)^{1/x}-x^{1/x}}{(x+3)^{1/x}-x^{1/x}}$$
I found the limits of each term become 1, which makes the fraction $\frac{0}{0}$, so I tried using L'Hospital's but it doesn't really help as the derivatives become even more complicated. I tried rationalizing either top or bottom, but it's not too easy either because the exponents contain $x$. I also tried using Squeeze Theorem, but all failed. I think there is a way to rationalize this to simplify and then use L'Hospital's from there but I really can't find a way. I don't normally give up but as I've been struggling with this for a month, I think it's time to seek help from others :( According to graphing devices, it looks like the limit approaches $\frac{2}{3}$. I want to find a way to verify this only using elementary calculus (no series expansion, or etc) as I think that's how Stewart intended.
Write the ratio as $\frac {(1+\frac 2 x )^{1/x} -1} {(1+\frac 3 x )^{1/x} -1}$. Thus you have to find $\lim_{y \to 0} \frac {(1+2y)^{y}-1} {(1+3y)^{y}-1}$. Can you take it from here?