If the equation a_0x^n+a_1x^(n-1)+...+a_(n-1)x = 0 has
If the equation $a_{0}x^{n}+a_{1}x^{\,n-1}+\cdots+a_{n-1}x = 0$ has a positive root $x = x_{0}$, show that the equation $na_{0}x^{\,n-1}+\bigl(n-1\bigr)a_{1}x^{\,n-2}+\cdots+a_{n-1} = 0$ has at least one positive root smaller than $x_{0}$.
Explanation of Method
The absent constant term supplies a second root for free.
Step 1: Name the polynomial
\[ f(x) = a_{0}x^{n}+a_{1}x^{\,n-1}+\cdots+a_{n-1}x. \]
Differentiating term by term reproduces the second equation exactly: $f'(x) = na_{0}x^{\,n-1}+\bigl(n-1\bigr)a_{1}x^{\,n-2}+\cdots+a_{n-1}$.
So the claim is that $f'$ has a root in $\bigl(0,x_{0}\bigr)$.
Step 2: The hidden second root
The polynomial has no constant term — every term carries a factor $x$: $f(x) = x\Bigl(a_{0}x^{\,n-1}+a_{1}x^{\,n-2}+\cdots+a_{n-1}\Bigr) \;\Longrightarrow\; f(0) = 0$.
This is the observation on which everything turns: the hypothesis supplies one root $x_{0}$, and the shape of the equation supplies another at the origin.
Step 3: Apply Rolle's theorem on $\bigl[0,x_{0}\bigr]$
$f$ is a polynomial, hence continuous on $\bigl[0,x_{0}\bigr]$ and differentiable on $\bigl(0,x_{0}\bigr)$; and $f(0) = f(x_{0}) = 0$. All three hypotheses hold, so there exists $\xi\in\bigl(0,x_{0}\bigr)$ with $f'(\xi) = 0$.
Step 4: Verification
\[\text{Take } f(x) = x^{3}-4x:\ \text{roots } 0,\pm2, \text{ so } x_{0} = 2\]
\[f'(x) = 3x^{2}-4 = 0 \;\Longrightarrow\; \xi = \frac{2}{\sqrt3} \approx 1.1547 \in \bigl(0,2\bigr) \;\;\checkmark\]
\[\text{Take } f(x) = x^{2}-3x:\ x_{0} = 3,\ f'(x) = 2x-3,\ \xi = 1.5 \in \bigl(0,3\bigr) \;\;\checkmark\]
Result
Final answer:
\[\boxed{\,\exists\,\xi\ \text{ with }\ 0<\xi<x_{0} \ \text{ and }\ na_{0}\xi^{\,n-1}+\bigl(n-1\bigr)a_{1}\xi^{\,n-2}+\cdots+a_{n-1} = 0\,}\]
Remark
Remark. The result is a disguised form of Problem 353's generalisation: between two roots of $f$ sits a root of $f'$. What makes the problem look harder than it is, is that only one root is named; recognising that the missing constant term forces $f(0) = 0$ turns it into a one-line application of Rolle's theorem.