# Where does the notation $\pi_1(X,x)$ for the fundamental group first appear?

I've spent the last half hour browsing Stillwell's translation of Poincaré's Analysis Situs and Dieudonné's History of Algebraic and Differential Topology, and I haven't found the source of this notation. The "1" in the subscript would lead one to believe this notation was introduced at the same time as $\pi_n(X,x)$ for the higher homotopy groups as well.

In my search, I have learned that the name "Fundamental group" comes straight from Poincaré, and from the "Fundamental region" of a group action. (And, incidentally, that "torsion" in algebra comes from the homology of non-orientable surfaces, which are twisted in on themselves. A bit of a tangent, but I'm still marveling.)

-
I would guess that it originates from Hurewicz's papers ("Beiträge zur Topologie der Deformationen I-IV") introducing higher homotopy groups, but I have been unable to find an electronic verson of them to verify this. –  Eric Wofsey Sep 25 '13 at 6:12
Small remark: your comment about Torsion was previously mentioned on MathOverflow by Qiaochu Yuan here: mathoverflow.net/a/13090/11540 –  David White Sep 25 '13 at 10:51
Possibly notation $\ \pi(X)\$ predated $\ \pi_n(X),\$ and $\ \pi_1(X)\$ in particular. –  Włodzimierz Holsztyński Sep 25 '13 at 15:22

Notice that in Poincare's setting there is no base point $x$, since he regards the fundamental group as a group of permutations of fundamental regions -- though he also knew the path interpretation, of course, as it arose in integration on Riemann surfaces.
The first occurrence of the notation $\pi_n(X)$ for the $n$-th homotopy group of space $X$ is Hurewicz's paper
If you want the base point $x$ as well, then I have failed. It seems not to appear anywhere in Math Reviews before 1945 but I did not find when it did occur.