%I A037257
%S A037257 1,3,9,20,38,64,100,148,209,284,374,480,603,745,908,1093,1301,1533,
%T A037257 1790,2075,2389,2733,3108,3515,3955,4429,4938,5484,6069,6694,7360,8068,
%U A037257 8819,9614,10454,11340,12273,13255,14287,15370,16505,17693,18935,20232
%N A037257 a() = 1,3,... [ A037257 ], differences = 2,... [ A037258 ] and 2nd differences 
               [ A037259 ] are disjoint and monotonic; adjoin next free number to 
               2nd differences unless it would produce a duplicate in which case 
               ignore.
%C A037257 27 and 250 are the first two numbers to be ignored.
%D A037257 I discovered this around 1979; Martin Gardner described a version of 
               it in "Weird Numbers from Titan", Isaac Asimov's SF Magazine, May 
               1980, pp. 42ff.
%e A037257 After 1 3 9 20 with differences
%e A037257 ------ 2 6 11 and 2nd differences
%e A037257 ------- 4 5, the next free number is 7 so we get
%e A037257 ----- 1 3 9 20 38 ...
%e A037257 ------ 2 6 11 18 ...
%e A037257 ------- 4 5 7 ....
%Y A037257 Cf. A005228, A030124.
%Y A037257 Sequence in context: A073716 A037048 A139142 this_sequence A145068 A027114 
               A145070
%Y A037257 Adjacent sequences: A037254 A037255 A037256 this_sequence A037258 A037259 
               A037260
%K A037257 nonn,easy,nice
%O A037257 0,2
%A A037257 N. J. A. Sloane (njas(AT)research.att.com).
%E A037257 More terms from Larry Reeves (larryr(AT)acm.org), Sep 25 2000