# On the number of subrings

Dear All,

It is well-known that a group is finite iff the number of its subgroups is finite. Is there any similar result for rings and subrings ?? (Clearly we have to distinguish two cases:

1. rings have identity and subrings have the same identity.

2. rings and subrings in general.

-
There is a third case, namely rings have to have identities but subrings don't necessarily share them. –  Qiaochu Yuan Dec 28 '12 at 9:07

As pointed out below, the result does not hold for rings with a unit as witnessed by $\mathbb{Z}$ which has no proper subrings. Also, $\mathbb{Z}[1/2]$ has two subrings including itself and $\mathbb{Z}[1/6]$ has four. This shows that there is an infinite ring with $2^k$ subrings. What other examples are there which are commutative integral domains? I don't immediately see any. I think this is a ring with $3$ proper subrings: $\mathbb{Z}[x]$ where $x^3=0\ne x^2$ and $7x=0.$ Of course $7$ can be replaced by any other prime.
The article seems to be referring to the second case. The result is clearly false in the first case, e.g. because $\mathbb{Z}$ has only one unital subring. –  Qiaochu Yuan Dec 28 '12 at 9:08