*Holding a line to read multiple records for one of the variables;
data a; 
input y @; 
* An ARRAY statement may handle this more efficiently;
do i=1 to 3; 
input x @; 
output;
end;
drop i; 
datalines; 
1 3.5 4.1 6.3 
3 4.2 3.3 6.6 
2 4.2 6.6 3.3 
; 
proc print; run; 
* Using both @@ and @ to read multiple x for a single y;
data a; 
input y @; 
* An ARRAY statement may handle this more efficiently;
do i=1 to 3; 
input x @ @@; 
output;
end;
drop i; 
datalines; 
1 3.5 4.1 6.3 3 4.2 3.3 6.6 
2 4.2 6.6 3.3 
; 
proc print; run;