Related to the question Condition for perfect packing of ellipse with circles along the major axis
An attempt of "perfectly pack" a parabola $x=ay^2$ as a special king of ellipse with eccentricity 1 results in a packed sequence of circles
\begin{align} r_0&=\frac1{2a} ,\quad r_1=3r_0 ,\quad r_2=5r_0 ,\quad \dots ,\quad r_n=(2n+1)r_0 \tag{1}\label{1} , \end{align}
where $r_0$ is the starting circle, internally tangent to parabola at its vertex, which agrees with the curvature at the vertex.
Question: Is there any known reference where this property of parabola is mentioned/discussed?

There is a paper that came out in 2019: Giovanni Lucca, Integer sequences, Pythagorean triplets and circle chains inscribed inside a parabola. The PDF is here.