1# nernernini

data a1;
note1=1;
run;
data a2;
note2=2;
run;
data a3;
note1=1;
run;
data a4;
note1=2;
note5=0;
run;
PROC SQL;
SELECT memname "datasets including variable 'note1'"
FROM DICTIONARY.COLUMNS
WHERE LIBNAME="WORK" and UPCASE(NAME)="NOTE1" and substr(memname,1,1)='A' and 0<input(substr(memname,2),5.)<=4
;
QUIT;

多谢jingju11。  
我原来是想用：varname 函数的。比较麻烦。还需要循环。你的这个方法很简捷。
再次感谢。