Sums of 5 or more positive squares Dickson [1] credits Dubouis with publishing
the following theorem in 1911. An integer n ≥ 34 can be written as a sum of k positive
squares for all k satisfying 5 ≤ k ≤ n except for k = n − 13, n − 10, n − 7,
n − 5, n − 4, n − 2, n − 1. Writing 20 years later, Pall [7] laments over having duplicated
Dubouis’ work before noticing the report of it but resists presenting his own
proof. Writing over 75 years later still, I suspect that both Dubouis’ and Pall’s proofs
resembled the following.