Conditions are obtained for (*)E|S T |γ 2 whereT is a stopping time and {S n=∑ 1 ,X j ℱ n ,n⩾1} is a martingale and these ensure when (**)X n ,n≥1 are independent, mean zero random variables that (*) holds wheneverET γ/2<∞, sup n≥1 E|X n |γ<∞. This, in turn, is applied to obtain conditions for the validity ofE|S k,T |γ<∞ and of the second moment equationES 2 =σ 2 EΣ S 2 where $$S_{k,n} = \sum\nolimits_{1 \leqslant i_t< \cdots< i_k \leqslant n} {X_{i_1 } ,...,X_{i_k } ,n \geqslant k \geqslant 2} $$ and {X n , n≥1} satisfies (**) and $$EX_n^2 = \sigma ^2< \infty $$ ,n≥1. The latter is utilized to elicit information about a moment of a stopping rule. It is also shown for i.i.d. {X n , n≥1} withEX=0,EX 2=1 that the a.s. limit set of {(n log logn)−k/2 S k,n ,n≥k} is [0,2 k/2/k!] or [−2 k/2/k!] according ask is even or odd and this can readily be reformulated in terms of the corresponding (degenerate kernel)U-statistic $$U_{k,n} = S_{_{_{k,n} } } /(_k^n )$$ .